It's not (just) about trust, it's mostly about how the CA certificate is "obtained".
Firefox supports only HTTP for retrieving an intermediate CA certificate, and can only do this "live".
Say you have a "leaf" certificate, and you already trust its Root CA, but there is no copy locally of the intermediate CA certificate. What now? Firefox will look for a HTTP URL in the leaf certificate pointing at the intermediate CA's certificate endpoint. It'll download and cache it on the fly.
This is fine... on the public internet with public CAs.
It's not generally robust on internal networks. For example, many "secure desktops" deployed in the military, security agencies, parliament, etc... don't have unfettered outbound access. They may need to talk to internal HTTPS endpoints but have no access to Internet-hosted CA endpoints publishing the intermediate certificate!
Windows solves this by allowing a single machine to download the CA certificates and the associated CRL files, and pushing this out to the workstations via policy. Firefox solves this by timing out and failing.
Similarly, HTTPS is the new-kid-on-the-block for PKI infrastructure! Its original use-case was for things like Smart Cards. Windows started out deploying certificates via LDAP, not HTTP! Firefox just does... nothing... if it sees only an LDAP URL in the certificate and you get an error.
Last but not least, truly paranoid admins don't just trust every public CA for every purpose. It's possible for a Windows network to deploy updated trust lists that restrict CA purposes. Similarly, it's possible to push out certificates you distrust and have anything using them rejected, whether or not the parent CA has published a CRL entry to revoke the certificate.
Right now, there's a lot of admins in Ukraine adding Russian CAs to such rejection lists. Get it?
Firefox has it's own unique and special way of managing all of this, and just ignores whatever the operating system tells it... except for a random subset here and there. That's not documented. You have to go read (and understand!) the C++ code to figure out what it does or does not do.
Oh, and to add injury to insult, the method for managing the Firefox certificate trust list changes semi-regularly, so most of the URLs in the docs are now 404s.
I've given up, and so have millions of other admins.
Firefox had a philosophical opposition to supporting enterprise customers on Windows, and they've paid the price for that philosophy with increasing irrelevance of their core product.
PS: Chrome / Chromium / Edge have their own issues, but nowhere near as bad. One major concern is that for non-public CAs used at a large scale, it's basically impossible to revoke certificates for any Chromium-based browser on unmanaged networks. It's still very difficult on managed networks. An example would be a University with tens of thousands of unmanaged workstations using an internal CA for things like VPN or WiFi access.