usually in live settings on rock\punk\metal band there are 2 guitarists, one plays the rhytem and the other one the lead, how is it conveyed in recorded material?
i'm interested in that matter becouse i never find my recording quite enought "full",and i'm wandering if some big album productoin really are made be 2 guitar parts( and single bass,vocal,drum track for each) for a total of 6-8 parts.
in other words, do you think that the richness and the "hook" of the song(prodution vise) come from multiple parts ( lets say 12-20 parts) or from the right mixing and mastering?
Not necessarily different parts being played, but most recordings in metal and hard rock are double tracked, sometimes quadruple tracked. Meaning, each rhythm part is recorded twice, each lead is recorded twice. Sometimes parts are played up an octave as well. So you might have a chorus that is power chords, those are double tracked, then the same chords are played an octave higher and layered into the mix.
You almost always want to double track the rhythm guitars, I don't tend to bother with quad tracking because there's too much potential for phase issues without any really noticeable payoff. I would only double the lead if I was doing it differently somehow (e.g. layering a solo up/down an octave or something) because when you double track you want to be panning either side, and you will notice it. With lead parts you usually want them in the center, so you don't want/need the double.
Also the secret to awesome, huge, fat guitars in a metal mix is something many metal guitarists like to pretend doesn't exist, it's the bass. Once you've double tracked the guitars put a high pass somewhere between 100-200Hz depending on how it sounds to you (I usually go about 180). Then create two bass tracks, on one of them put a low pass around 300-500Hz (again depending on sound). On the other, grab some distortion plugins and get it sounding really nasty dirty and midrangey, then highpass between 300 - 500Hz, and lowpass between 2 - 5kHz. Set the dirty track to a volume where you can just hear it through the guitars, then back it off a little, and bring up the low end track until it sounds good.
Don't overdo leads to the point where you can't do it live easily. Like listen to Muse's song Hysteria. Great song but there is a point where there are 4 separate lead parts going on plus a rhythm guitar. Obviously not impossible live but Muse is a 3-piece in studio and add a Keyboardist live.
Derpy Derp Derp Herp Derp
Lots of good advice above.

If you can play the solo the exact same way every time, don't be afraid to double-track the lead. You can put both parts straight up the middle as you typically would with a single lead track. Randy Rhodes did this all the time. The thing is, if you can't play them practically identically, it'll just be a mess. In other words, your solos need to be prepared, not improvised.

With doubling, though, you get to a point where you're actually defeating yourself. It's easy to make the leap from "that double-tracking sounds great" to "so dodeca-tracking would sound absolutely incredible." No. You will get to a point where all those tracks, rather than helping each other stand up, actually just fight with each other and start to sound less. You'll get to a point where actually getting RID of some of those parts makes the mix open up and sound better.

This is a blog I put together on creating space and depth in a mix:

Could I get some more talent in the monitors, please?

I know it sounds crazy, but try to learn to inhale your voice. www.thebelcantotechnique.com

Chris is the king of relating music things to other objects in real life.
A touch of reverb on the guitars can also make them sound a little bigger, just be warned with reverb comes clutter in the lower frequencies and can be muddy. All advice above is great, especially the point about bass - it can bring a whole song together, and really fill out the song. Use the stereo field! Properly panned elements can make the sound sound wide and huge (also drums are a big part of it). EQ your instruments, chatterbox272 said it perfectly.
Most good stuff has already been said, I'll just add two other things that apply mainly to rock/metal.

The biggest thing I always find people forgetting, thanks to bands like Pantera's mixes, is the MIDRANGE. Scooping the mids may sound cool in isolation or when playing along to CD's in your bedroom, but put a guitar with hardly any mids in a mix and it will be crushed by everything else so all you end up hearing is a thin, fizzy sound from the high end that makes it through the drum overheads. When tracking guitars, especially through a big beefy 4x12 driven by an all-valve amp, I usually set the 'Low' as low as 3 (it's not very important in the mix and can increase undesirable, boomy resonances in the cab when palm muting); the 'Mid' up as high as 7 or 7.5; and the 'High' can vary anywhere from neutral (5, or "12 o'clock") to 6.5. You can sculpt the sound further with mic placement and tweak that EQ to suit your sound when placing the mic/evaluating the sound in the room, but I find that's a good starting point for me to work from.

The other thing is when people crank up the gain to make things easier to play, or because they think it's "more brutalz", without realising it is reducing the clarity of everything and lessening the impact by spreading the transients through distorted artifacts and widening the 'plateau' of each section of the waveform (because the top of each wave is lopped off, creating a wider, flatter 'peak' level which sounds especially terrible on many solid state amps, but even valve amps can sound horrid when the gain goes too high for the style you want!). As a general rule, I would tell people to start with as little gain as you require to play the song accurately and cleanly, and then add the gain for tone's sake, rather than ease of playing. Of course, that depends on the skill level of the player, but if you can't play most things on a lightly overdriven guitar, you probably can't perform them flawlessly live either - some things obviously need more gain than others (whammy harmonics, pinch 'squeals' and fast legato across multiple strings etc.) but if you're writing music with complex sweep arpeggios and a tapped note at the top, you should at least be at a level where you can judge how much gain you really 'need'.
Hey, look. Sigs are back.