Fwiw, as you may know, I don't give a toss about modeling being 100, 95, 81 or even just 61% true to the original.
And I will as well happily accept that there's still certain differences (whether that's for good or bad shouldn't matter here, both exist).
But it's completely beyond me how some people can come up with weird snakeoil-ish assumptions, followed by some "that's why analog is better" resumes. And as we were at it, claiming that modeled amps would always compress more than their analog counterparts is just that, namely non-provable voodoo. If they were doing this, it could be proven easily. And heck, some defenitely do, such as the GT-1000 amps (just that they don't exactly have analog counterparts). But coming up with some sweeping generalizations is just stupid, simply because it's not true.