Here's the catch though; You aren't gonna go anywhere with any of these services if your game selection is not up to snuff, and as it stands it still won't be up to snuff with almost any of these services. It's hard to compete with the 30,000+ games of Steam.
For now, the final business model is still undecided and there's two factions at NVIDIA for GeForce Now, but one of the two that is currently winning inside the company is running Steam directly on their infrastructure, with your personal account. That's what the current GeForce Now beta client currently does.
The other faction leans on the "netflix of gaming" side.
Still I don't think it matters nearly as much as you think it does. Nintendo wouldn't be able to sell boatloads of switch otherwise. Google only needs a decent selection of video games to make it work, they don't need the full 30k of steam.
The rest of your comment is a reflection on the current market which is typical shortsighted, static views of current instead of what could be. Nobody wanted a smartphone (Nokia communicators, Windows Mobile) before the iPhone. Now everyone owns a smartphone, although not necessarily an iPhone. With the right timing, technology and vision new markets cam form, new audiences can be brought in. If you go solely by revenue potential mobile market and its cazul shit games is just as big as the games you've cited. Heck, Fortnite is popular on mobiles, that should tell you something about its audience and willingness to put up with things like unsuited control schemes. A little latency wouldn't be a killer so long as it stays within reasonable amounts. Once again I'm repeating myself but you don't realize how fearsome the idea of being able to play a game directly from clicking a button on youtube after watching its advertisement trailer can be.
This will either spectacularly fail or become a next huge market in competition with what already exists.
Your point about US datacaps and other US shenanigans is valid though. The US won't be a targetable market for the years to come. That may be something that will change, but it'll take time.
For starters, 5G is an impossibility if the US doesn't improve its fiber infrastructure and pull so much fiber everywhere that it wouldn't make sense to not go for the last mile and make use of it for residential internet too. So while things won't change
tomorrow I expect the US situation to get better in the near future, considering how much cell phone providers are attracted to 5G.
In France, fiber deployment has finally gotten really serious. Unless you live real deep in the boonies you'll have some form of high speed internet, if not now, then pretty soon.