The infinite injury article is (IMO) complete hogwash. Its author assumes (but never explores or explains the assumption) that there's some unidirectional arrow of improved understanding and pedagogy through time (“progress”) and that therefore modern explanations are inherently more sophisticated/subtle/polished than the original grapplings with a subject.
What he misses is that the seminal papers in any field are forced to really grapple with a subject for its own sake, while later works fetishize dogma and ceaselessly pander to the interests of whoever is paying for the work to be done, simplifying concepts to be understood by those of uncertain prior experience, and packing in features (ooh, glossy pictures! companion website!) that will sell copies.
There are some areas where studying the originals is not especially helpful, because subjects weren’t yet understood. For instance, the founders of quantum mechanics really had no clue what was going on at the beginning, and their early papers are mishmashes of math pulled from other sources which sometimes fortuitously explained their experimental results. So by all means, go grab Townsend’s Modern Approach to Quantum Mechanics (2002) as a more practical and informative work.
In other areas, the famous “originals” (which is to say, those works which were good enough to last centuries while their contemporaries faded into oblivion) are wonderful. There’s much more insight packed into Machiavelli, Tocqueville, or James Madison’s writing than any modern political science textbooks.
And yes, economics as a field would be way better off if its practitioners had any idea what its foundational assumptions were about, or tried really deeply considering (e.g.) Smith, Veblen, Marx, or Keynes (or even, say, Von Neumann and Morgenstern), who had a real world to answer to, not just other economists.