Interesting. I am not sure that I agree with any of this, but I also admit that for running a 5k, I am not sure that I can point to a very specific physiological reason why runs much longer than an hour are important. I do think that they are important; but I really cannot point to anything beyond the general tenets of improving muscular and connective tissue strength, increasing number of mitochondria in muscle cells, increasing stroke volume in heart, etc. All of the things that you generally get from easy distance, but get more of from more easy distance.
As to how long folks should keep working base miles, I find that one to be interesting from a personal perspective. I was a college walk on with no running experience and got pretty decent pretty quickly my freshman year, but had a real breakthrough my senior year.
Most of the period from about 23 through my late 30s were spent running on and off, but with no consistency and more time off than on. I could get into decent shape off of a few months of running (low to mid 16's for a 5k) off of that. But just before I turned 40 I got pretty serious about it again and began running consistent mileage most of the year round, with some exceptions for planned breaks between training cycles.
When I was up at my heaviest volume at age 43 (getting ready for a marathon), I ran my best times in years (sub 15:30).
You can argue that the high mileage was the key to my improvement. On the other hand, you could argue that I was really rebuilding a base over those three or four years, just as I had built a base from my freshman to senior year of college before breaking through back then.
Either way you look at it, it seems that 3-4 years of consistent training was a benchmark for me in seeing material improvements. Whether I could continue to improve, or improve more, by reducing the volume and upping the intensity at that point is an unanswered question.