@Brian, the lag copy - if I understand correctly - does not really solve the issue. The only step of the ridiculous process we have to go through to restore a mailbox that the lag copy avoids is the restore of the database file from backups. We still have to do the recovery DB, export the PST, somehow get the user the PST, have them dig through it to get what they want, then tear it all down again.
The lag copy simply saves a bit of time (most of which is just waiting for the restore itself, not active working time) - and it means that we have the space of the entire database taken up at all times instead of just during a restore event.
It really is ridiculous not to have proper item-level restore in a product as advanced as Exchange - my clients are continually baffled when I explain to them the way it works currently.