Like everything else in the modern world, the good and progressive idea of "open access publishing" has been turned into a corporate kāmadhēnu. This list probably needs no background on the general issues (see
FOASAS), but it is totally standard for journals to charge "article processing charges" for open-access publication. I don't know what you mean by "our area of studies," but certainly journals like
Indo-Iranian Journal, Journal of South Asian Intellectual History, and
Philological Encounters (Brill, $2,595),
Journal of Indian Philosophy (Springer, $3,2900),
International Journal of Hindu Studies (Springer, $3,190),
Indian Economic and Social History Review (Sage, $3,000–$4,000), and
South Asia (Taylor and Francis, $3,300) do charge very high APCs for open-access publication. (They don't charge APCs for traditional articles, but as we know, the subscriptions make these APCs look cheap.) Journals published by university presses (like
Bulletin of the School of Oriental and African Studies), I believe, negotiate APCs on a per-case basis, and I don't know what the average is. According to
Wikipedia, the average APC is now $1,626, which makes
Religions technically on the cheaper end, and I believe that MDPI is technically a non-profit. But MDPI was classified as a "predatory publishers" by Jeffrey Beall in the past, and still today their core model involves (a) volume (they are the largest open-access publisher in the world), (b) speed (anecdotally,
Religions has a faster turnaround rate than almost any other journal in our "field"), and (c) special issues. The last feature, which you also mentioned in your message, is connected with a number of intellectual and ethical issues, including (a) the relative roles, and relative amount of labor, of the journal staff (including the editors) and the organizers of the special issue in soliciting papers, organizing review and copyediting, helping the authors revise, etc. (b) the integrity and the rigor of the peer review process.
The same, by the way, is true for monographs, where the open-access fees are generally between $12,000 and $18,000.
Now obviously I think this situation is unfortunate and wrong, and I believe (as someone on Reddit said) "friends don't let friends publish in MDPI," but this is unfortunately the way that academic publishing is developing, or rather has developed. Still, I would assume that in the case of most, if not all, of the special issues you linked to, the APCs for all of the papers were probably paid by the editor's home institution, or through a funded project. That was the case when I published in JSAIH recently. And it may be the case that the editors are able to broker some kind of deal with the publisher. (I now see that Brian Collins has refuted these naive assumptions of mine --- I was trying not to be so cynical!)
The choices for researchers in our field, especially early-career researchers, remain pretty miserable. If you want to publish your research in a major journal, the probability approaches 100% that it will be owned by one of the "big five" corporate publishers, with huge profit margins, or alternatively, pay-to-play schemes like MDPI. There are few truly independent journals (I believe
Bulletin de l'École française d'Extrême-Orient is one), and few "diamond" open-access journals, which do not charge APCs (see the
FOASAS again), but which are mostly just starting out and do not have the name-recognition, selectivity, or even the indexing that researchers (and promotion committees) often look for.