The focus on problem solving may be good for helping kids learn and practice solving problems procedurally. The free form nature of scratch gives a lot of freedom for goofing off, which can be a good thing.
Scratch is great and noble and better in every way as an isolated programming learning tool, but it's worth noting the reach and marketability of Pocky to get kids interested in the whole scene.
You (parents, kids, non-CS-oriented teachers and mentors) aren't going to find and use Scratch without the prime directive of "I want a kid-friendly programming learning tool".
But now you can get the kids Pocky, which you may have gotten anyway since no kid is going to complain about that, and they can learn something meaningful and cool with a character that has context they know and appreciate, unlike scratch's characters that exist in their bubble. Actually being able to impact a much larger audience even with a less ideal experience, is net worth much more than a stellar tool that cannot break past a high activation barrier and caters to a very tiny audience. Like Settlers of Catan for introducing people to board games outside Monopoly, and IFLS (I Fucking Love Science) for getting people interested in space, biology, etc. These seem like perverted contractions of beautiful things, but at the end of the day they get more people interested and that's a good thing.
I for one remember countless mornings reading and learning the backs of the cereal boxes as I had breakfast getting ready for school, and those mornings would have been so much more productive building Pocky programs, but there was no way I was going to fire up Scratch and create programs there.