The corollary of "never say never" is "don't always say always" BUT when I'm mix (in post) live recordings, I do always buss all the instruments (except bass, usually) together and put a neutral/colorless compressor on that buss side-chained to vocal(s). Now, I don't always use it, but I will typically flip it on and off and touch the threshold and compression levels (keeping it on fast attack and release), and listen. This can keep the vocals from getting trampled on right at the places where it's trying to get through, like on the initial consonants and parts where the vocals get louder, but too typically (especially with less experienced musicians), the music gets even louder!
The other thing to play with (or even try first as it can be simpler) and listen is using a mix knob on your vocal compressor if it has one. This lets you try parallel compression, which might do something similar, in letting some vocal dynamics survive the compressor, but still do some squeezing.
I use Logic's Space Designer for reverb, and as [MENTION=22368]Tadpui[/MENTION] suggests, have it highpassed and lowpassed. I generally (but not always!) have a tiny bit of pre-delay on everything (~10ms) unless I trying to push something back and then I might set up another reverb send with 0 predelay.