What are you trying to show there? That video is running with no RT which makes it equivalent to the PS5's Performance mode. As a reminder, this is what the PS5 runs at in it's different modes:
Quality: 1584p-4k 30fps, RT
Performance: 1440p-4k 60fp, no RT
Performance RT: 1080p-1440p, 60fps, RT
The PC in that scene is running at a fixed 4K DLSS quality which means internal rendering resolution is 1440p but with the added overhead of DLSS. So the actual performance will sit somewhere between 1440p and 4k. The PS5 using DRS on the other hand can drop as low as 1440p internal and then use the less performance intensive IGTI to bring it back up to 4k. Given that is a particularly intensive scene it stands to reason that the PS5 would be running at or near it's lower DRS limit there and so it effectively would have an upscaling performance advantage over the PC (with inferior image quality as the trade off).
On top of that, the PC is not at lower settings as you suggest, but higher. This can be checked from 12:10 onwards. The PC is using Vey High Textures and High Hair settings just like the PS5. It's also using 8x AF and HBAO+ AO which are both higher than the PS5.
Add to all of this that this mode is running without RT which is where the 3060 is able to stretch it's legs vs the PS5 (in fact we should expect a 2070 class GPU to be a bit slower than the PS5 where RT is not involved) and it's clear that this is falling far short of your earlier claim that the 3060 (to say nothing of the roughly 12% faster 6600XT) gets "outperformed cleanly in all of the modes".
The other sections of that video can't be readily compared to the PS5 because not only are they using the higher base graphics and upscaling settings mentioned, above, but they are also using Very High RT resolution and geometry.
