I guess it would be helpful to somehow "map" the stages that determine female and male development (or none). I've tried to do that below based on my basic understanding of the issue. If anyone wants to build on it, or correct me, please do!
Step 1
So I guess it starts with the SRY gene (normally located on the Y chromosome, but could be present on the X chromosome).
+SRY = tells the gonadal tissue to develop into testes - Male pathway
-SRY (plus other factors??) = gonadal tissue develops into ovaries - Female pathway
Step 2
Then whether there are any other factors which stop the SRY gene from functioning correctly (that 'block' it, or I guess receptors in the body aren't functioning correctly).
SRY gene not blocked/ receptors functional - Male pathway
SRY gene blocked/receptors not functional - Female pathway
Step 3
Then you've got whether the testes actually function correctly
Testes produce testosterone - Male pathway
Testes don't produce testosterone - I don't think this leads to a complete female pathway as T is converted to oestrogen, so T needs to be produced? I guess this would lead to an androgynous presentation with XY genes and feminised gonads?
Step 4
Then you've got whether the testosterone is converted into DHT (the "more potent" version of T)
DHT produced - Male pathway
DHT not produced - Assuming presence of T, this is a Male pathway with feminised appearance (until puberty - as with the 5-ARD)
Step 5
Then you've got whether the testosterone receptors are working:
T/DHT received as normal - Male pathway
T/DHT somehow not received. - Female pathway, although PAIS I guess there might be some male pathway.
I'm sure there are other steps that go wrong but these seem like the ones covered in the discussions above. Also sure that this isn't necessarily a linear process!
From what I gather as well, there are a couple of key developmental steps - In utero development, a few weeks postnatal development, then puberty - which have a subtle role too. For XX with no DSD (i.e. women!), an exposure to T in utero can be a strong indicator of 'male characteristics' including a longer ring finger in comparison to the index finger, higher likelihood of attraction to females, and a higher likelihood of "masculine" behavioural traits. This is not considered a DSD.
Male homosexuality and "feminine" behavioural traits does not seem to be caused by lower T during any of these stages.