Current state: /dev/md0 (comprised of 8 drives in a RAID 6 array XFS file system)
Future state: /dev/md0 (will be adding 4 more drives so a total of 12, RAID 6)

mdadm --add /dev/md0 /dev/sdk /dev/sdl /dev/sdm /dev/sdn
mdadm -v --grow --raid-devices=12 /dev/md0

And since I'm using XFS, after the sync is done, execute: xfs_growfs /dev/md0

Anything need adjusted in the fstab file? (I'm not on that PC now to check but if it's something obvious, like any references to 8 drives and switch it to 12, I can handle. But beyond that, any 'gotchas' I should watch out for?

PS The array is a data drive. My boot device is a SSD. Speaking of, once this is all said and done, is it a wise idea to keep a copy of the mdadm.conf file in the cloud or somewhere in case I need to reinstall Linux? (Disaster recovery and all.

I want to make 100% certain that's the right way because if I screw up the array and lose it, I'll need to restore
14+ TB of data and that's gonna take about 20+ hours of non-stop copying.

Whenever you work on a raid array, you should have a backup. The main purpose of raid is in the event of drive failure, you don't need to restore from backup. You can just pop in a new drive without taking the system down.

But you still need backups! Raid is not a substitute for backups. It's a way to avoid downtime. Your above commands look fine. I don't think you need the xfs_growfs command.