Ok, tank you. Here's a follow-up question:
I never tested MIR or MIRx until present, so my assumption could be wrong ... I imagine MIR Pro as a big box that collects the audio streams from all your VI tracks, does some math to it, and returns you a stereo mix of your whole project, hence it is "all or nothing at all" - either your machine is capable of handling all the audio data, or it isn't. There is no use to "freeze" individual tracks including the MIR reverb, because it is always the whole big MIR engine running - no matter how many tracks you use.
As MIRx is built into each instance of VI you simply can turn it off while you are still working on your project and use some standard Reverb on the main output of your DAW. When it comes to mixing, you turn on the MIRx-Button in all your tracks and bounce-in-place them one after another, which gives you the opportunity to use the MIR technology on slower machines.
Or am I wrong and MIR Pro works in a different way?