The 100$ fee per transaction is what it would cost to sustain current miner revenue without mining reward (ie. mining revenue derived somewhat invisibly from money supply increase).
The piece is not garbage, but a thoughtful introduction to a problem that's approaching inexorably.
I'm sure miners would love to see their revenue never decrease, but this isn't a problem for users or for the security of the network. Block reward does not need to remain at or above current levels forever; it just needs to remain predictable from block to block. Difficulty adjustment works both up and down, even if we've only seen it go up so far. The 51% risk is overblown - the network is significantly overprotected right now, and we could stand to lose a few orders of magnitude without jeopardizing security/availability. Plus, it might even placate the energy consumption doomsayers.
As GP notes, empty blocks are now common, which is in fact a bigger risk in a low-subsidy environment. Maybe we should instead reduce minrelaytxfee to 0.1 sat/vB to incentivize full blocks.
The piece is not garbage, but a thoughtful introduction to a problem that's approaching inexorably.