Depends what we mean by "its job". I'm sure all of Vanguard's funds are doing exactly what their prospectuses (prospecti?) say, but what the author means is that you don't get more diversification from having 3600 market-weighted stocks vs. the S&P500. On a more theoretical economics level, the point could be made that the efficient frontier is not expanded when you already have 500+ stocks and then add a couple thousand more. Thus there are diminishing returns to diversification (an already well-known fact).
I think the article might have been more clear if it said something like "buying VTI doesn't sufficiently diversify investors from the S&P 500 because large caps are so exponentially bigger than small caps that a market weighted fund is still 96% large-cap." and "investors who want to hedge against the possibility of lower multiples for large tech companies, stagnation of large companies, or a resurgence in value investing should consider specific ETFs".
Investors worried about particular narratives, such as Apple being disrupted by a Chinese competitor, Amazon being disrupted by a new distributed server technology they can't monetize, laws making companies like Microsoft liable for hacks of their software, or antitrust action against any of them, could be interested in buying an S&P 600 or Russel 2000 fund. Of course, doing so involves discounting the competing narrative that these giant corporations will take over the economy and push out the small-caps. It also involves discounting the narrative that PE ratios need to fall back to earth, because small-caps have historically had very high PE's.
https://siblisresearch.com/data/russell-2000-pe-yield/