Hotter iron or more patience for stubborn solder on a Pi Zero?

Fresh leaded solder on top is absolutely the move, I've done it a dozen times on Pi boards. The stuff already in those holes is almost certainly lead-free, and that stuff melts way hotter than what most of us are used to. Mixing in some 60/40 lowers the melting point of the whole blob and suddenly it actually flows. Keep your iron at a normal temp, like what you'd use for any regular joint, touch fresh solder to the pad and let it mix with the old material, then drag it out with braid while everything's still molten. What I'd avoid is cranking the heat higher. I learned that one the hard way years ago when I lifted a pad off an ESP32 board trying to brute-force it with temprature. The heat just doesn't get where you need it through that big thermal mass anyway, since the ground plane on those boards sucks it away, so you end up cooking the top surface while the hole stays solid. A chunky chisel tip helps way more than raw heat because it just transfers more energy at a sane setting.

Also, if a stubborn bit refuses to move, heating the pin hole from the top side while feeding solder from underneath can break it loose. On the Zero there's room to get at both sides if you've got the board propped up. Flick the melted blob out with tweezers if wicking fails, tap the board edge on the desk while it's liquid and it'll often just fall out.

Honestly, I'd question whether the holes need to be empty in the first place. When I put right-angle headers on my Zero boards, I never strip the pads clean. I put flux on, melt each blob with a decent chisel tip untill it's properly liquid, then push the pin straight in through the molten solder and hold it for a few seconds while it sets. The old stuff just gets displaced around the pin. Works every time and the joint ends up solid. Chasing a perfectly clean hole on a double-sided board with ground pours is how you cook pads for no reason.

The trick is thermal mass, not more scrubbing. A fine conical tip will sit there touching the blob and losing the fight. My station runs about 380 with a 3mm chisel, and I touch the pad and the pin at the same time so heat goes into both. Feed in a little fresh 63/37 as you go and the whole mess turns over fast. If you really do want the holes empty, an iron plus wick on a full plugged hole is miserable. A hand pump the second the blob goes liquid works far better, and a cheap hot air setup with some foil masking over the nearby parts is even less fuss. But for a header swap I'd just melt and insert, done in ten minutes.

I do the solder-in + braid method but with one tweak that made a huge difference for me: I run my iron at 375 degrees for this kind of cleanup instead of my usual 330, and I freshen the braid tip constantly. Burnt braid stops wicking almost imediately and just cakes the flux, which is prolly the 'skins over again' thing you're seeing. Cut a fresh half inch of braid every couple of holes. Cheap and it keeps things moving fast enough that the ground plane never wins.

I've had decent luck warming the whole board up first when solder's being stubborn like that. A heat gun on low from the underside for a minute or so takes the chill off, and then the iron doesn't have to do all the work by itself. Everything goes liquid way easier and you're not leaning on one spot forever trying to get heat through.

Also don't sleep on a solder sucker for holes like this. Melt the blob, jam the nozzle over it, hit the button. Gets most of it out in one shot and the braid only has to mop up the rest.

One follow-up on the cleanup itself: even after the blob is gone, the hole often still has a paper-thin film of solder clinging to the plating inside, wich looks clear but will stop a pin from seating flush. While it's still hot, poke a sewing needle through from the top and you'll feel right away whether the path is actually open. Then clean the flux off with isopropyl and give the pads a look under a magnifier, because a half-lifted pad is easy to miss until you're staring at it with the header already soldered in.

Not buying the 'lead-free melts way hotter' bit from the first reply. It runs maybe 30 or 40 degrees above 60/40, that's it, it's not some super alloy. The reason those blobs feel impossible is the ground plane under every GPIO pad drinking the heat away, plain and simple. Mix in fresh solder if you like, it does help the flow, but don't credit the temperature difference for it. On mine I just sat the iron on one hole for a solid few seconds longer than felt comfortable and the whole thing let go. Patience beat temperature.