Washington ? The Columbia investigation board did not go far enough in its recommended safety changes, one of the investigators says in a supplemental report that urges NASA to strengthen shuttle inspections and correct mechanical problems that were unrelated to the disaster but could cause another.

Air Force Brig. Gen. Duane Deal said Wednesday he felt compelled to highlight these issues after they ended up being buried, downplayed or dropped from the final report of the Columbia Accident Investigation Board.

“I feel an obligation that if I know of something that could cause the next accident that’s waiting to happen and I didn’t bring it forward, that’s when I wouldn’t be able to look myself in the mirror,” Deal said in an interview with The Associated Press.

Deal stressed that his 10-page supplement, which will appear in an upcoming volume of appendices, is not a dissenting opinion. It started out as a minority opinion a week ago, he acknowledged, but many of the 12 other board members jumped on board.

“We are all very proud of this report,” he said of the recommendations made public Tuesday. “We think it does, in fact, what we wanted it to do.

“Some people think it ought to do more, some think we’re too blunt. But I think there’s almost universal agreement that our 207 days of work were good.”

But given NASA’s reputation for ignoring reports, Deal said he was skeptical the space agency would fulfill all 29 recommendations in the full report, let alone the ones referred to as observations.

Deal, for example, worries that NASA will give short shrift to the report’s 10th chapter, titled “Other Significant Observations.” The observations include corroding shuttle parts, brittle bolts that support wing panels, failures in the system that releases the shuttle from the pad at liftoff and weakened rings attaching the fuel tank to the booster rockets.

None of these problems, but rather a piece of flyaway fuel-tank insulating foam, caused the Columbia accident, which killed all seven astronauts aboard.

Instead of observations, some of the items should have been recommendations, Deal wrote, and one — the need for an independent bottoms-up review of Kennedy Space Center’s shuttle safety inspection protocol — should be a recommendation carried out before the next shuttle flight. As it is now, inspectors must justify to managers why certain critical parts should have mandatory checks rather than justify why they should not, he said.