What are the advantages and disadvantages of Preprints.org compared with arXiv?

You should generally use the service that most people in your field uses. In my subfield of physics people follow the daily postings on arXiv, so that's the natural choice for me. I imagine a preprint would be effectively invisible if posted elsewhere, except to someone who happens to be browsing my Google Scholar profile, ResearchGate profile or website. One could perhaps post to both arXiv and Preprints.org for wider dissemination, but that might confuse e.g. Google Scholar's citation counts so I'm not sure if it's a net benefit.

Advantages of Preprints.org

• Much wider field coverage.
• Can assign DOI to preprints. (Given that arXiv identifiers play essentially the same role, I personally don't think this is a large advantage, but some people seem to have a preference for DOIs.)
• Has a comment system. (The closest thing on arXiv would be emailing the authors.)

Disadvantages of Preprints.org

• No flexibility with license. All Preprints.org preprints are posted under a Creative Commons CC-BY 4.0 license. This makes some sense as Preprints.org is designed with open access journals in mind, but really limits the set of journals the manuscript can be published in.
• It's not clear to me how Preprints.org would handle misleading comments. One might prefer not to have those on the same page as the preprint.
• It's run by MDPI, a publisher that's been rather controversial in the past. Although they claim that Preprints.org is run on a not-for-profit basis, it remains fully funded by MDPI, which may or may not be a sustainable and lasting investment. (Compare e.g. the discontinued Nature Precedings.) In contrast, arXiv is operated by Cornell University (which at least I consider a more respectable entity), and has proven to have lasting power.

Unknowns

• One can volunteer to screen preprints on Preprints.org. Is that more or less robust than the screening by arXiv moderators?

If the use of arXiv in your field of science is common, that is probably the better option for you personally. For example, subject specific abstract indexing services (like inSPIRE or ADS for high energy physics and astronomy) will automatically link their entries for the arXiv version to the journal version. Consequently, you will easily be able to obtain combined citation counts.

However, there are some limits to arXiv. One of the foremost is that they accept pre-prints only for a limited number of subject areas (essentially physics+astronomy+mathematics+(some) computer science with maybe some cross-over into other fields). Consequently, for some fields arXiv simply is not an option. (There also some subfields for which arXiv would accept pre-prints, but the use of arXiv is not so common). In this case, alternatives could be interesting.

Personally, I have never heard of preprints.org. One potential worry I would have is that it is owned by a (commercial) publisher. So, even though it is currently run as a non-profit with free access, I am not sure what safeguards there are against them changing their usage policy in the future. I would certainly try to find out before submitting anything.