It has been 5 days since recovering my hard disk. Too soon to be certain, but long enough to make a preliminary assessment.
smartctl still says the Current_Pending_Sector and Offline_Uncorrectable counters are both zero. It also says the last recorded error is the one from the original problem - no new errors have been logged.
I am puzzled that the raw counters for Raw_Read_Error_Rate (e.g. 204765846 -> 93045057), Seek_Error_Rate and Hardware_ECC_Recovered are significantly large, but actually lower than 5 days ago. I wonder what caused them to reset - or perhaps the binary counters just overflowed and wrapped to zero?
I will reserve judgement on whether the disk needs replacement until I have more data... Thankfully I can now watch all my recovered programs and offload decrypted versions of them to safety!