I have an orchestral template that has served me well. But I have two doubts that I wanted to know if I could improve:

1. Reverb:
   1. According to the image below, I simulate the equivalent of two mics. One mic near the instrument and another mic to get the reverb from the room. So, for each instrument, I send the dry signal directly to Master Bus, plus a Send to a reverb Bus (Altiverb). Because if I send all only to Altiverb, I lose control of pan and depth (actually I use the Virtual Sound Stage for this). 
   2. Also, I use Pre-Fader / Post-Fader, depending on whether I want to leave the microphone more or less "distant", to have more or less room sound or direct sound.
   3. In summary, I have 2 outputs for each instrument: 1 dry for pan control and another as send for reverb control. Is this the best approach?
      
2. Often all this setup sounds good, but sometimes there are, for example, melodic passages of a clarinet or any other instrument that is more "distant". In this case I have to bring the instrument forward (automate the volume knob with pre-fader). My question is: in a traditional orchestra mix, is it common to create automations to bring an instrument forward at the solo passage and then return it further back in the other passages of the music?

A "traditional orchestra mix" is created by the conductor on the soundstage as the music is being played and recorded by a stereo pair of microphones.

Lead guitar out front.
 
 

bdickens
A "traditional orchestra mix" is created by the conductor on the soundstage as the music is being played and recorded by a stereo pair of microphones.

What? I do not know if you understood my questions. Still, I think you're wrong about the microphones:

My doubts persist.

Heck, I just experiment with pan for each inst, use a little verb on each synth, then put Spaces (Northeastern Hall) on the master bus. When I say experiment with pan,  I mean I've put hundreds of hours into trying different arrangements. I'm sure you can do better with fancy (and expensive) toys like Mir, but you can also do a lot with careful attention to pan, reverb and of course the various volume adjustments.

I use EWQLSO and to be honest, I don't feel any of the tracks need a lot of mixing and quite often use them as they are, straight out of the box in terms of EQ, Pan & Reverb.
 
But I do use all the options provided by the Play engine for room & close mics

I've been using EWQLSO for 10 years, but nowadays it's very limited compared to the new libraries.
Currently I use a large set of libraries, such as LASS, Sample Modeling, Berlin Woodwinds, Vir2 Elite Orchestral Percussion, etc.
But as each library has a different behavior and sonority, I'm forced to make complex adjustments.
And to increase realism, rather than just panning, I use Virtual Sound Stage 2:

 
Repeating my initial questions, I would like to know:

1. Is enough just one individual dry signal for each instrument (including the FX Virtual Sound Stage) + one send to a reverb bus? Or is there anything that can be done better?
2. Is it common to create automations to bring an instrument forward at the solo passage and then return it further back in the other passages of the music?

Most orchestral libraries are already pre-panned according to the traditional stage arrangement, so further panning isn't necessary - if, of course, you are content with the traditional form. Hollywood film scores don't conform to tradition, nor do pop/rock orchestral backing tracks, so there is plenty of precedent for violating tradition. Personally, even though I tend to stick to conventions, I have no ethical qualms about panning basses to the center.
 
Front-to-back positioning, however, is usually up to you. That means adjusting send levels to the reverb bus for each section to simulate their distance from the listener.
 
Whether you're going full classical or modern Hollywood, there will be times when you do have to position an instrument yourself. For instance, in an acoustical performance a solo instrument will normally move to the front and center of the stage. In that case, you'd pan it center, reduce its reverb send, boost its treble and turn it up. (Don't forget about the role of EQ in creating the depth illusion.)
 
My biggest challenge is with reverb. Because I often mix and match libraries, it takes some trial and error to bring dry samples' reverb up to match the more ambient samples while maintaining the front-to-back illusion. Sometimes I have to resort to multiple reverbs if the disparity is great.

Hi, I do a lot virtual orchestration - both mixing and writing and for me I use one verb, with starting settings of -3 into the buss for the front orchestral instruments, and -6 for the back. And I use one of those orchestral placement charts too. To create the spatial feel of an orchestra, on top of the above reverb settings - I use a Wave's plug called the Brauer Motion plug...in fact it is 29 bucks at the moment. This plug allows me to individually place stereo audio tracks, by shrinking the stereo field or widening it - depending on the instrument within the ideal orchestral placement chart. This means you don't need to do any fader moves, apart from the odd instrument buss, particularly if you have set-up your dynamics right when you're writing the piece in whatever score editor your using. Of course, if you're only midi sequencing...it's harder to use my above formula. I also use virtual binaural plugs on the verb buss...this formula creates a very realistic  virtual orchestra. There are a couple of other neat tricks to get my sound...tape sims, and stereo widening, but I'm not going to reveal all my secrets to the special sauce. 
 
Ben     

bitflipper
 (Don't forget about the role of EQ in creating the depth illusion.)
 

Thanks again, Dave.
On the subject above, I did not know. Do you have any links for some explanation or video tutorials?