I think it would go without saying that for most crops, having more sun and better machinery access would be superior. But the farmer couldn't make it viable without the solar. Once they had solar, the produce was another positive factor.
Similar to solar roadways, I would expect there are many reasons to separate the solar panels and the farm.
For example, maybe the solar panels restrict the types of tractors you can use. Maybe the fields sometimes create too much condensation and harm the solar panels. Maybe pesticides damage the solar panels. Maybe the solar panels sometimes block too much sun from reaching the plants. And so on...
They're not arguing to have found a global maximum and everybody everywhere should stop what they're doing and adopt this scheme, this is just reporting that there are unexpected positive effects to the scheme they've tried.
I'm just pointing out a common engineering principle: Trying to optimize two solutions together is often more difficult than trying to optimize them separately.
I hope I'm not coming across as overly negative here. If these things work together, that's very cool! But I would still be curious to see an overall comparison between doing them together vs doing them separately.