Whether you believe this or not comes down to whether you believe references 32-34 from the paper. I haven't read the stuff about transgenic expression of human ace2 in mice, but it would sound a lot more like it was aberrantly expressed, so I'm not counting that. "More evidence needed" is the correct interpretation here, I believe.
There's other literature out there saying ACE2 is ubiquitously expressed (including brain) - so that means one of two things: People saying it's expressed everywhere are wrong, or SARS-COV-2 requires more than ACE2 for infection. There are papers saying that transfecting cell lines with ACE2 doesn't always render the cells susceptible, so it's probably a good idea to consider the requirement for a co-receptor.
Like anything, the true answer probably lies in between all these possibilities.