Niklas Luhmann became one of the most productive sociologists of the 20th century with the help of his enormous paper-based Zettelkasten. If you look at the stuff he wrote, you can see why. He ties together publications from the fields of sociology, philosophy, legal studies, psychology, biology and surely many more, literature, journalism, film... Luhmann was a prolific reader (he did few other things as far as I know) and for him, stumbling upon a connecting thought he had ten years ago while reading a newspaper after having read a specific book, might have been crucial to maintain the density of ideas in his publications.
In short, these tools are probably only useful to you if you're in the business of generating novel ideas by interlinking a lot of other ideas that people have had in new, interesting ways. (This is the best tentative description I came up with and it's probably wrong around the edges).
If you're an engineer, or indeed also a scholar in the humanities but playing a different game than Luhmann, these tools may just be useless to you. A couple of years ago when I was thinking about this a lot, I asked one of my lecturers who was a post-doc in comparative political science about his toolset. He didn't really seem to understand the question, he told me that he sometimes writes notes on books in a Word document but mostly knows what's going on in his field and where to look for what. I later took a look at his dissertation and while I'm in no position to judge the quality of his work (it was probably pretty solid, he got it published with a reputable publisher), it seemed to have fever moving parts and threads of thought tied together than the bits of Luhmann that I've read.
My impression is that what's holding the ecosystem of tools for networked thought back right now is that the tools are not built for (or possibly even by) the people for whom networked thought may be most useful. They're trying to be better task managers, tagging systems, collection managers (as mentioned in the linked article), flashcard systems, etc. "Zettelkasten" by Daniel Lüdecke (a sociologist), the software recommended in "How to Take Smart Notes" by Sönke Ahrens (himself a professor in the humanities), which is hailed as the bible of networked thought by many (Roam, Logseq, HN I guess), looks very different from these tools. It's an obscure piece of Java software, and while I've only briefly tried it out for a few minutes, it works very differently. Smaller notes, little structure within them, no titles. It has a "desk mode" where you can pull out notes and arrange them in a tree structure for when you're writing a paper or book (see https://www.youtube.com/watch?v=XIztPpFqCBw).
I would love to see what things could happen in the networked thought space if people who need these tools the most sat down with people who can write software. I have a few ideas, but I'm afraid I'm just not enough of a Luhmann to really know what these people need. Also, I would just be procrastinating actually writing my thesis (nod to the linked article :))