But I don't unzip archives, I use a reader like mcomix or zathura.
For example: I'd enter the first image, swipe left and suddenly I was looking at page 5 and the next image would then be page 3.
So, my best guess is that whatever image viewer you're using is sorting images by local file creation time, which might be out of order with pages if the translators updated page, or maybe if your unzip process finishes writing files out of order. Do the order of the filenames not match the page order?
E.g. for the link I gave, the files unzip as 000.png through 013.png, which defines a sequence, which matches the page sequence, and my image browser ('feh') is smart enough to go by the filenames, even if I change the file modification times. But a different one might sort the images by time.
Also feh is using lexographic sort, not numerical; if I rename 002.png to 2.png, it shows up at the end.