I will have a go at this when I get home this afternoon.
But my initial 2c, and please don't take this as moaning, but rather trying to be constructive and drive improvement: modelling pre-amps and power amps to match hardware has been done before, notably the Lecto tests of a real amp and Lecto running through the same real cab / mic set up were so close that a few minor tweaks of the amp EQ wouldn't fix.
Where the lowest fruit for next leap in innovation lies is in the cab emulation. You've taken that a step forward with the dynamics knob on Recabinet, but there's more to go. In high gain metal, where it's really the bite / bark of the guitar (2k + Hz) that's accentuated, this isn't really a problem. But when you go to a more full, earthy tone, there's a static low-mid drone when you pan the guitars hard L-R, that really sucks the stereo spread of the guitars.
When you hear tracks of real amped guitars (or even Kemper), it sounds there is two independent guitar tracks and there's no common low-mid drone dragging them into the centre of the stereo field.
Over the next week I'm trying to organise a Kemper reamp, a Nebula reamp and my own IR based mix to give an example of what I'm talking about (I can't afford a proper re-amp, but feel free to volunteer anyone), and hopefully spark a discussion about the difference. If there is no difference, then I will admit I'm full of sh*t and go back to my corner.
