That's more or less what I'm talking about here. I think the recent revisions have been focused on professionals. But a language's long term growth depends on new people being trained in it to replace the old programmers who are retiring.
I agree, but then again "C++" has no budget to allocate, it depends on contributions from people who are not paid anything to contribute to C++ or companies that are concerned about issues that make their developers more productive. If some billionaire got group of highly talented devs and they worked on a lot of stuff you want(wrote prototypes, wrote nice proposals) I doubt C++ standardization would reject them. But I do not see something like this happening soon.
So I hope I am wrong, but I see very little from your list getting fixed in next 5y(except maybe networking).
Making C++ easier to teach can overlap with making it easier to use for professional developers. For example, part of the motivation for P1679 was to make C++ easier to teach, though it will also make the language more pleasant to use for experienced developers.
Before P1679:
if (haystack.find("needle") != std::string::npos)
With C++23:
if (haystack.contains("needle"))
C++20 also added .contains() to containers and starts_with()/ends_with() to strings. Making C++ more natural to read lets developers focus more on the task being solved than the language details.