I am uniquely qualified to respond to this, as about 6 years ago I developed a raw milk cheese that was inspired by Stilton that eventually won 1st place in its class in the World Champion Cheese Contest.

I do not feel that the original claim is overstated. Creating any novel cheese is remarkably difficult–seriously, it's really damn hard–but recreating a cheese that no longer exists and was likely made in a different location under different circumstances is effectively impossible. There are so many hidden variables involved that, even if they had access to the model cheese, a talented cheesemaker could still take years to make a good approximation. And it would only be an approximation. Even with today's advanced cheese technology you cannot recreate terrior.

Since they don't have access to the model cheese, this difficulty is compounded many times.

Edit: I want to add that even though I was at the peak of my cheese game, I was only able to do this because I had effectively unlimited money to throw at the problem and very little oversight from my investors.