That example you mentioned would be just a couple of lines of html with css (which you could inline if you wanted), and there are plenty of drag and drop tools for the web (WebFlow is a modern example, but they have been around for decades).
And there are plenty of other “layouting primitives”, all of which are quite cumbersome. And due to HTML encoding some form of semantic meaning only (though seldom used even for that), drag n drop tools can’t generate easy to read/modify output. Also, without js frameworks, you can’t really just embed another component into it trivially.
And I’m not trying to shit on web tech, JS has absolutely overcome its shitty past, and is quite a likable language nowadays, and CSS is an extremely powerful and expressive system (though this latter is due in part to how shitty html is). My only gripe is with HTML that is simply not fit for the task (as it was never meant for that).
If by structure you mean things like hbox/vbox then no, html is not supposed to do that. HTML should only represent the semantic structure.
There are a couple of exceptions for interfaces that are genuinely unconventional, or that deal with multiple spacial dimensions, or that are primarily representing graphs and visual/audio data. But most interfaces should be representable in pure text.
This argument that "the web is only for documents" is missing the point that most applications are just interactive documents when you really break down their UI. A labeled tree is a fine way to represent user-facing state for most apps, and most apps would benefit from exposing a more semantic interface to users that was easier for them to manipulate and read programmatically.
Swing, Qt, et al do not have great answers to this problem, and definitely not superior ones than css.
but I agree in general with your statement.