As far as building a wall of sustained and distorted electric guitar, try to give eack guitar track as individual a sonic personality as possible. The basic mixing rules of wantng to fill and balance the three dimensions of the soundstage still apply; they don't just go out the window because we're talking all guitars.
This means stuff like when recording your individual tracks to be assembled into the wall like bricks, using different pickups, mixing up the stomp box chain a bit, using different amp channels, switching microphones, and using some tongue-and-groove EQing all to get the tracks to play together spectrum-wise. Recording the exact same guitar sound over and over again will result in a lot of mud; even though it's a wall of guitars, there has to be some sonic mortar seperating the individual bricks, otherise it'll just be a messy pile of bricks and not a nice solid wall.
This also means balancing these sonic difference in pan space. For example, if you have, say, six guitar tracks and tracks 1 and 3 sound almost the same frequency characteristics-wise, seperate them in pan space left and right (not necessarily hard panned; while you can have some gits hard-panned, not everything with 6 strings need to be hard panned, even in metal.) Layer your guitars across the wall so that not all alike-sounding tracks are in one spot; seperate the fuller-sounding tracks with brighter-sounding tracks and vice versa.
Your arrangement will dictate what's best compression-wise once you're mixing. While most people individually compress every track to the max and then stack them, when building a wall, sometimes it's best to ease off on the amount of individual track compression, adding just enough to thicken the track just a little, but not squash it. Then when the tracks are assembled into the wall, add the final amount of smaller compression to the entire wall (perhaps as a submix) to give it some overall glue. You'll have to figure out which way is best for your composition.
HTH,
G.
