The biggest thing I've taken away from all of my reading and experimentation in this (and being a bit of an information hoarder myself) is that the link itself, or the article itself, is not information, it's "data."
There is value to the data, and if you want to record it that's fine. I use tags (hashtags in Logseq) to classify data so at least it's in some type of taxonomy, but Logseq full-text search is also pretty good so as long as you have words around the URLs that you'd be likely to search for you're good.
Information is what you derive from the data. It's your summary, or paraphrasing of the core point(s), or connections you make between a thing and some other thing you previously recorded.
Tools like Logseq and Obsidian exist to allow you to easily create those connections, and that's very much based on the zettelkasten method, though amplified by what the technology now allows, which is more complex and nuanced.
Don't hoard URLs and headlines. Hoard your thoughts about them, what you think is important about them, what you saw that related to another thing you saw. That is information.
https://zettelkasten.de/introduction/