sure, but there's a limit to that....
the minimum output size (i.e. the smallest amount of possible BTC you can send) is currently 182 satoshis (and that's segwit outputs only, non-segwit outputs have a minimum size of 546 satoshis)
so that means things will change somehow before we get to 2140, otherwise the value of BTC will be in effect capped to 182 satoshis being equivalent to the smallest individual item one can buy (so maybe the price of a single nail in a hardware store? something like that)
But the idea always was that 1 BTC could continually divide into smaller denominations to reflect the increasing buying power of the currency. It seems now that 2.1 quadrillion (i.e. 21 million BTC multiplied by 100 million satoshis constituting each 1 BTC) is not going to be enough, whether BTC is a dominant currency or not. Milli-satoshis are already in use on the Lightning network, but I see Lightning as a medium term stop-gap to some other way of solving the various scaling issues. Maybe Lightning itself will evolve to becomes an all-encompassing solution with no compromises, but it doesn't quite seem like that just yet (I'd happily be proven wrong however).
So there's an unknowable scenario in store for Bitcoin; only once all innovation in efficient production of goods has ceased will we know how many decimal places we need to subdivide each satoshi into. And only then will we know what the economics of mining with only fees as a reward will be. I guess we could speculate on the range of possibilities, but that naturally includes different ways of mining and/or structuring the blockchain (innovation is innovative ).
But of course, none of this will matter for at least 100 years. There's alot to do between now and then without trying to tackle these issues seriously, maybe something that happens between now and then will make this sort of speculation entirely a moot point.