your blocks ends up increasing the risk that you get orphaned since nodes prefer the first block they heard.
I think this assumption of theirs is the flaw.
Successful pools do not build on the
first block they hear; they build on
the most difficult block they hear.
You're wrong: nobody does that and doing so puts the miner at a disadvantage because the block you hear about first is the one most likely to have propagated to a majority of the network. It is however a possible way to mitigate this attack and in an email to bitcoin-development one of the authors specifically stated they thought of that idea and left it out of their paper due to space constraints:
Here is a solution we did not put in the paper due to space constraints
that should alleviate your concern:
Instead of locally choosing a block at random, have a deterministic
pseudo-random mechanism for choosing between competing chains. E.g., take
the one whose last block hash is smaller. This way all miners choose the
same chain, and the guarantees of our solution hold.
I'm working on analyzing a stronger version of this solution that would make the choice to mine the smaller block hash be short-term economically rational for miners.