Well, the plan has always been to eventually bundle at least three, and up to nine or more, rockets together to achieve launch capabilities far in excess of a single rocket without dramatically increasing the cost by building huge specialty rockets. And if they go up together, they'll need to come down together (well, anything that doesn't make it all the way to orbit)
There are some serious challenges with the proposal, but mostly with the "going up" part of the equation. If you can land one autonomous rocket with the precision currently needed* , while hitting as impressively close to the bullseye as they've been doing, then having something else doing the same thing nearby probably won't be a huge issue.
On the other hand, I wouldn't be surprised if by "simultaneous" they mean "within minutes of each other" so that they don't have to maintain formation during the entire landing process. As long as they can avoid knocking over the previously landed rockets with the backwash that would seem to introduce a lot fewer opportunities for something to go wrong.
* they can't throttle down the rockets to be weak enough to hover, so they have to hit close-enough-to-zero speed and altitude at exactly the same moment, and then cut the engines before they start going back up again.