Hi - The most common use for compressors in a mix is on vocals, and not just to level out screaming and quite parts. Even if a line is sung at a consistant level, there are peaks and dips in the level that when mixed in with the music will make the vocal alternately pop out, and get burried. Compression is invaluable in making the vocal sit in the mix. Sometimes an uncompressed vocal is what the song calls for, but it's rare.
Compression ratio's very, depending on the style and the desired consistancy of the vocal, and the consistancy of the singer. 4:1 is still on the light side, with 10:1 giving a really flattened track. The other interactive factor is the threshold setting, which determines the level at which everything below that level is left alone. Set it while watching the gain reduction meter, to either just smooth the peaks, or to squash the entire signal, or somewhere in between. Much experimentation is called for, and it may be different for every song. Oh, and maybe try a little faster attack, with a slow release. Slow attack times can give a very audible breathing efect that is not desirable.
Cheers, RD