From my interpretation you don't account for the devices/footprints being non-symmetrical along both axis to make this work.
Simple devices like resistors or non-polar capacitors can do it - symmetrical on two axes.
Diodes are only symmetrical to one axis.
Anything else usually is not symmetrical at all.
This means, if you 'mirror' the arrangement of the devices, the tracks need to go to other pins - it can't just be mirrored.
The only time you can 'mirror' the layout without problem is when you FLIP it to the backside (if you had been on the front before, or vise versa). But that means all the devices/tracks that had been on the front are now on the back and vice versa. It's not exactly a mirror, but looks like one if looked at from the top (or bottom).
That's why @maui mentioned this visual board flipping option in the opengl canvas.
But this won't be applicable to real world hardware..
Do me a favor. Take an IC, any will do.
Place it in front of you on the table on a sheet of paper. Note where it's pin #1 is.
Place marks for the pins on the paper, noting pin #1 as well.
Then MIRROR the footprint on the paper how you think this should be working.
Now TRY to place the IC onto that footprint, with pin #1 matching the pin #1 on the mirrored footprint..