You're going to feel more difference - subjectively - with any of the usual STBs than you'll actually gain in terms of chassis stiffness.
Chassis torsional stiffness is a problem in 3 dimensions, and the best that any 2-point STB can do is close in the chassis structure at a single longitudinal location. You do see this stiffening effect gradually diminish over perhaps a few inches either side of the STB location, but it's still correct to consider this stiffening effect to be localized.
In other words, you're stiffening only a single vertical-transverse plane along the car's entire wheelbase, and if you're effectively stiffening, say, 6" of length out of a Mustang's 107" wheelbase, you're not increasing the stiffness of the other 101" at all. Adding a diagonal brace in that same vertical-transverse plane does make that plane stiffer, but still won't do anything for the other 101".
Both the Garage Time guy and the Engineering Explained guy note that when you tie the two towers together that you get to use both towers' stiffnesses. Which is true as far as it goes, but does not tell the whole story (or even the most useful part of it). I think I'm going to have to draw some pictures before it'll make sense, but it has to do with the fact that the inside and outside tires are not loaded equally and do not produce equal cornering forces.
Garage Time guy's measured results weren't the least bit surprising. A different kind of 3-point brace would most likely have shown considerably more improvement.
Back to the subjective part . . . one thing that even light-duty stiffeners can do is change the structural vibration picture. They can chase more noticeable vibration modes out of the picture completely, where you're left with less noticeable modeshapes. You have to be visualizing the car structure vibrating like a massively complex tuning fork here, that when you tie off a point that's vibrating a lot relative to another point to that other point you change that vibration shape completely. A little like grabbing the tips of a simple tuning fork between your fingers, actually.
Vibrations are a part of what you perceive as "chassis solidity", which isn't the same thing as chassis stiffness (though there is a loose relationship between the two). That's why it 'feels' better, the car doesn't feel as "loosey-goosey" as it might have before. This concept can and should be extended to many of the other add-on stiffening that's available.
Sketches are going to take some time. Maybe I'll include a sketch suggesting the possible benefit of running a 2-point STB specifically for hard braking . . . and I have at least one thought related to stiffnesses in general.
I've been down this road before (a number of times, on various forums). As a structural analyst/engineer, this kind of stuff is very much in my wheelhouse. I'm retired these days, but I didn't suddenly forget everything I knew about the concepts.
Norm