The Flows have low impedance but also low sensitivity/efficiency - which is really just a fancy way of saying they need power
There's a handy post here which helps you calculate how much power headphones actually need:
https://nwavguy.blogspot.com/2011/09/more-power.html
the AFC and AFO both have nominal 13 ohm impedance, the AFC's efficiency is 92dB/mW, the AFO is 94dB/mW. From the table we can see the AFCs are going to need somewhere around 60mW power to drive them to 110dB peak SPL (which is a very high level, but nwavguy recommends it as a good guide for various reasons, e.g. recordings with high dynamic range where the average volume may be 30dB below the peak). That's not crazy high, but it is higher than IEMs and headphones specifically intended for portable use.
Looking at the AK70's specs, it looks like it ought to be able to do that, though:
https://avprogear.com/astellkern-ak70-mk-ii-portable-premium-dap/
There's another handy resource here -
https://www.owenduffy.net/calc/voltcnv.htm - which lets you convert between all kinds of different ways of specifying power. The lowest specified power for the AK70 (mark 1, unbalanced) is 2.0Vrms. If you put 13 ohms and 2.0Vrms into the calculator, it'll show 307mW, which is more than enough, in theory. However, that output spec was measured with no load, which is a kinda flattering way of doing it. Amps can't always manage the same amount of output power into different input impedances. Fortunately, the reference-audio-analyzer folks measured the AK70 for us, so we can find out:
https://reference-audio-analyzer.pro/en/report/amp/astell-kern-ak70-unbalance.php
there's a whole buncha scary numbers there, but if you ignore most of 'em and go to the
"Maximum voltage output level vs load" table, it gets interesting. Look at the "Optimal or equal class AB" Vrms column (the whole 'class' thing is a bit complicated, see https://reference-audio-analyzer.pro/en/test-amp-class-power.php#ab , but it sorta-kinda means "max power before the audio quality may possibly be measurably compromised"), and we see it can only manage 0.17 Vrms into 16 ohm (close enough to 13 for rough figuring). Hell, even under "Max, before clipping" - which is 'absolutely max power it'll put out, regardless of whether the signal gets a bit noisy' - we can see it can only manage an absolute max of 0.27 Vrms into 16 ohm. It *can* manage 2.0Vrms - AK aren't lying - but it can only really do it into 763ohms, or no load at all.
So, let's go back to the calculator and try plugging in 0.17 Vrms and 13 ohm...and we only get 2.22mW. Even 0.27 Vrms (the 'max, before clipping' number) is only 5.6mW. To put it another way, if we scroll down a bit to the 'math part' of the post, nwavguy gives us this calculation to figure out exactly how loud a given source can drive a given pair of headphones:
dBSPL = Sensitivity in dB/mW + 10 * LOG ( Pmax in mW)
I *think* that's meant to read Sensitivity in dB/mW + (10 * LOG Pmax in mW) , otherwise the calculations come out weird. So if we plug in the numbers there we get 92 + 10 * LOG 5.6 , which if I didn't mess up the ordering, is 99.48dB - the absolute loudest the measured AK70 could make a pair of AFCs play (in fact it may be slightly lower, since the measurement was for 16 ohm not 13 ohm, presumably the real measurement at 13 ohm would be slightly lower). That'd be the *peak* volume on a track; the average volume would be lower (though for highly compressed music, not much lower).
So: yeah, it's not too surprising your AK70 is kinda maxed out driving the AFCs, based on those measurements at least; if you have a different model from the one they measured, or you're using the balanced output, the numbers would be different, but probably not crazy different.
The AK70's output impedance was measured at 3.40 ohm; that's not *very* high, and probably isn't causing a problem, but it is kinda borderline. I think the general guideline is that the output impedance should be max 1/4 of the input impedance, and 13/4 is 3.25, so yeah, right on the line. (edit: seems nwavguy recommends 1/8, so this *could* be a problem. There's more info on the actual problems caused by using an amp with a higher output impedance and headphones with lower input impedance at https://nwavguy.blogspot.com/2011/02/headphone-amp-impedance.html?m=1 . For Aeons the second point - "Headphone impedance changes with frequency" - doesn't matter, as for planars the impedance does *not* in fact change with frequency, but the other two points may be relevant.)
I mean, you say it's loud enough for you at 130/150, so is this *really* a problem? Do you sometimes need to play it louder? Just the fact that you're at a high number doesn't necessarily mean it's a problem. You're likely in the range where the measured THD+noise is kinda above the ideal level (the difference between 'class AB' and 'Max, before clipping' from earlier), but if you're not *hearing* any obvious issues, is there really a problem?
If you are worried, though, yes, you *could* get more power with an amp or a different DAP: look for one that can manage at least 0.9Vrms into 16 ohms, and ideally has <1ohm output impedance just for safety. There are plenty like that. (FWIW I'm currently hanging onto an Opus #1s and an iBasso DX200 and trying to pick which one to keep, both of those can power AFCs easily, but there are many other options too).