The way I do it is to tweak each track as needed, first. Generally, I start with noise reduction, then run the non-bass tracks through the highpass filter set at anywhere from 100-150Hz, depending on the instrument/voice to clean up the boominess that would otherwise build up from my room. If I use reverb, I generally apply it to individual tracks to start with. The reason for that is that reverb can make a track sound farther away. I also compress individual tracks -- I don't do overall compression until practically the last step before converting to 44.1/16 for CD burning.
Most singers think they sound awesome soaked with reverb, and pitiful without. This is not really a mixdown problem, but more of an interpersonal issue, when you are recording. Assuming you are adding reverb to vocals, you would want a light reverb on the lead, and heavier reverbs on the backing vocals. If you do it the other way around, the lead will sound like it's behind the backup. I have a room with a nice natural reverb, and I will occasionally add a little artificial reverb to backups or instruments in the background.
If you want an overall reverb to give a sense of a band playing in a room, then do the reverb on the mix. This can be done after you have added some to individual tracks, and will help everything "meld" together. Just keep a light touch. A little goes a long way.
Hope this helps.