love your stuff!
It sounds like the main acoustic track is already pretty compressed - is that right?
at :45, when the snare comes in, the other instruments (including the snare) wash out the vocals -- that's a pretty hooky vocal part, so I'd try and fix that.
Around 1:09 as the acoustic solo starts, there's this huge swoosh in white noise - is that the noise floor from all the various guitar tracks being consolidated and amplified by a compressor? It really doesn't sound bad, but it's distracting. The noise then abruptly disappears at 1:32 - again distracting.
I'd be proud if I did what you just did with this, so take any suggestions as merely suggestions, not criticism:
I think I hear the tell-tale HF distortion that I associate with low-end (e.g., Chinese - although there's plenty of higher end Chinese stuff these days) mics and/or preamps - I tend to associate it with ceramic capacitors in the signal path. It also sounds like you're mixing without much headroom.
If I was working on this, first, I would probably take a simple gain plug (comes stock in most DAWs) and knock about 10 or 15 dB off each track individually as the first thing in the chain to give myself a bunch of headroom. I would consider HP and LP filtering early in the chain as well. bx_cleansweep works great for both of these tasks and it's free, and pretty CPU friendly.
My goal would be to leave a bunch of room in the main stereo output bus so that I could try various bus compressors (and then probably not use any of them).
I'd probably try taking a noise suppressor plugin - like Waves NS1 - and put it on the guitar tracks somewhere. It's easy to destroy a track with NS1, but it can also work wonders - it tends to cut away at that HF stuff. Sometimes I put that plug on each track near the top of the stack, but sometimes it's more effective after some compression has brought up the noise. I bet there are some alternatives to NS1 - I just don't know what they are (and I would like to know).
Because I've been into it recently, I'd try a tape simulator on all the tracks and again on the output - my pattern has been to use UAD Studer on the individual tracks (or sometimes stems), and then UAD Ampex on the stereo output bus. There are a bunch of other ones as well. Be wary of adding noise, since you already have it
-- most of these plugs allow you to choose whether you want tape hiss and hum or not.
For me, either no compression or very light compression is best with acoustic guitars. I have luck with the opto or vari-mu emulations like
LA-2A or Fairchild. I would probably compress the main track (be it the rhythm in the beginning or the lead at the end) individually, and the others together in their own bus.
After playing with all that, there would hopefully be tons of headroom left in the output. Then I'd either 1) send it to a mastering person exactly like that, or 2) if I wanted to make a louder mp3, I'd use a limiter like Fabfilter Pro-L or the amazing and free (and CPU-murdering) Vlad G Limiter No.6. I like those limiters, and some others, but I always like what the masterer does better