I imagine for people unfamiliar with AD&D and older editions that the argument that "fighters need non-combat features" seems strange or possibly an attempt to homogenize class design.
Back in older editions, fighter followers were a big thing. Yes, in either 1e or 2e other classes got followers BUT...
(1) The numbers, equipment, and skill level of other classes' followers were never as impressive as the fighter, and...
(2) Becoming a baron was built into the fighter in older editions. They were the only class that had explicit ties to political leadership and the ability to gain tax revenue.
That unique aspect of becoming a baron was stripped away from the fighter in 3e, or possibly in later 2e, can't remember exactly.
You may argue "Well good! I never wanted to play a political leader!" and that's fine, but it doesn't change the fact that something was taken away from the fighter which gave it unique non-combat functionality.
And nothing took its place.
So that's where the argument comes from (an argument I highly support) that the fighter needs non-combat class features. And I think [MENTION=37579]Jester David[/MENTION] hit the nail on the head in his comment above that the best place for such features are the sub-classes.
There is truth to your statement about fighters. In 1E & 2E they could gain followers. In 1E they specifically mentioned that fighters could collect taxes as could clerics. Paladins could build keeps and collect fair taxes whatever that means because they could only have LG henchmen and followers. I imagine that their domains would be rather small due to this alignment restriction. Rangers were always written up as mobile and restless. Rangers as landed lords were generally discounted because 1E rules stipulated that at most three rangers were allowed to work together.