The CDC study being discussed in this paper didn't followup but according to OP Kleck's other studies did - in Kleck's studies when people said they had a DGU they were asked to elaborate on the circumstances so this could be evaluated. By his analysis about 9% of reported DGUs didn't qualify as defensive, so he applied that as a modifier when comparing the new results to his.

On the flip side there also exist reasons to think the CDC undercounts DGUs, most notably that they only asked people who had a gun in the home if they had used a gun in self-defense. If respondents only had a gun at work or used someone else's gun or used to have a gun at home but no longer did, they weren't asked the DGU question. Also, most DGUs are arguably illegal so reporting one is reporting having engaged in illegal activity.

If you want to make the data more accurate you don't get to just apply reasons to think it's an overcount, you also need to apply reasons to think it's an undercount. (On the whole, Kleck's estimates seem pretty conservative.)