...-in the creators opinion, for the tasks that it solves for them

I think it's difficult to read "the one programming language that's close to being flawless", without interpreting it as being someone's opinion.

Reading the article though, I'd say a good objective argument is made that Fexl is "flawless", insofar as it being the "thinnest possible functional programming layer on top of C." Whether that's useful or good is another matter.

These being equivalent:

  (a (b (c (d (e (f x))))))
  (a; b; c; d; e; f x)

Is not required so isn't the thinnest possible layer.

Whether it's preferable to a homoiconic s-expr is subjective.

It's hard to take seriously the style opinions of someone who indents curly braces that way.