I'm in dry northern CA, and toggling between different modes didn't make a huge difference. All it's doing internally is tweaking the Gain and Offset values (which are exposed in the E-COTI, but not the TAD). I agree that "Arctic" strangely gave me reduced thermal noise on a warm dry CA night, but again, the difference was minor. I should point out that toggling presets in High Contrast mode made less of a difference than in Outline mode. The latter is really improved in the TAD.
All of that unwanted clutter is really the biggest crux with these COTI-style fusion systems. The thermal sensors are quite small on these devices and they work within the 8-12 micrometer spectrum, but I don't think that's sensitive enough to isolate the extremes of that spectrum. So with I^2 NV you see an image that's basically a monochrome version of what you're used to seeing, with shadows and highlights. Add the COTI-style thermal and now you see a range of heat as well, but it's a range. Within that range will be a lot of things in the environment, which creates clutter that complicates that I^2 NV image and can give you a lot of false positives.
The Auto Brightness plays a part too, because it works like a thermal "exposure." For example, if I stop and stare into a patch of trees, that thermal clutter may settle a bit and anything warm within will light up. But then, if I scan upwards the sky becomes the coldest object in the sensor's FOV, so the Auto Brightness adjusts by lighting up the trees because they're warmer than the sky. I need to experiment more with Auto Brightness disabled, but previous experience doing this with the E-COTI didn't help the situation.
Either way, it's still a really powerful combination and the TAD is clearly a more economical option for us no-longer-MIL guys. My initial hunch that "great concept, but needs more technological evolution" may still be valid.