A couple of items worth noting:
Firstly, Valeton will be exhibiting at a trade show in Germany at the end of September. Their teaser post says this:
"What to Expect:
Hands-on access to our full pedal lineup—plus brand-new models!
Direct access to the Valeton team.
A sneak peek at what’s fresh off the factory floor."
That points at some new products being shown in production prototype form, just as the 150 & 180 were shown in October last year. It's consistent with the expected timeline for the GP300, although nothing is certain of course.
Secondly, you may recall from my original posts that there was circumstantial evidence for the GP-300 removing the traditional Valeton limitation of "just one effect of each type". Some additional indirect evidence in support of this comes from an unexpected quarter - a home-brew firmware for the GP150/GP180:
This firmware allows most effect blocks to be run with any effect type, which implies that the underlying official firmware makes that patch relative easy. We know that Valeton announced the GP150 & GP180 as having a new architecture, and initially this appeared to boil down to nothing more than a switch to 48kHz, since that was the only obvious change and the actual effect algorithms were unchanged.
This new info changes that somewhat. We know from the excellent GP200 reverse engineering effort (https://gp200-reversing.hashnode.dev/the-gp-200-landscape) that the GP200's internal architecture does not lend itself to being easily patched for multiple MOD instances: there is no indirection layer and so to have a chorus in two blocks the actual code would have to be duplicated. That this GP180 patch is able to have any effect in any block while still being a small patch implies that the 150/180 (and by extension, the 300) do have that indirection layer, which would mean that the 300 would easily be able to offer multiple effect instances as a differentiator and an upsell reason from the 200.
Now, I don't necessarily expect the GP300 to offer completely freeform effect arrangement, because Valeton may still decide to retain artificial limitations in the interest of stopping the user shoot themselves in the foot. For example, they may only allow a single instance of AMP and CAB, but enable multiple MOD or DLY blocks. Nevertheless, this gives additional grounds for hope, if not outright optimism.