The hexbright's body is the main thing. The insides could be completely new.
My plan is to keep it running. Perhaps replace the emitter by a more efficient one at some point, and repair anything as needed. If somebody bothers to make a better board someday, I will upgrade to that as long as it is still open hardware.
I think the Hexbright's body would be a source of complaints from present-day reviewers. It's bulky and heavy for a 1x18650 light; having to expose the bare LED to the elements to charge or change the battery is risky; anodized threads between the LED and outer body are bad for heat transfer.
LEDs with higher luminous efficacy than the XM-L2 exist now, but for a large increase, they usually have a larger light emitting surface. That means a less focused hotspot and less useful beam distance with the same power input. On the other hand, it's possible to achieve a more focused hotspot and more beam distance without giving up more output with some newer LEDs optimized for high output from a smaller surface. It's also possible to greatly improve color rendering.
My plan is to keep it running. Perhaps replace the emitter by a more efficient one at some point, and repair anything as needed. If somebody bothers to make a better board someday, I will upgrade to that as long as it is still open hardware.