If your employer uses Watchguard firewalls, they can proxy SSL connections, replace the SSL certificate with one signed by the firewall, and act as a MITM to filter and report the connection contents. If your employer has pushed out certificates by, e.g. active directory and group policies on Windows, then the site will appear padlocked without warnings and you will be none the wiser unless you inspect the certificate. I'm assuming other firewalling products can do similar.
This means if, unknown to you, CINIC impersonates a domain which uses SSL, and you visit that domain and assume your SSL connection will be safe from prying Chinese officials, you're incorrect. Incorrect because Chinese officials could have created that bogus SSL certificate, and it would be accepted by Firefox because Firefox now accept CINIC as a root authority.
Any confirmation, in case I'm spreading inaccuracies, would be appreciated.
As mentioned, CNNIC's inclusion only allows a Man In The Middle Attack. That is, CNNIC, i.e. the Chinese government, pretending to be a site they're not, and using their status as a Firefox (and other) -authorised root certificate provider to further trick the user into accepting the bogus SSL certificate as originating from the hijacked domain.
I am vaguely aware of MITM attacks: that someone sends you their public key while pretending to be someone else. And this means the data you send is encrypted in a way that the MITM can see.
However, I'm unsure how CINIC's inclusion in Firefox's root certificates facilitates this. Perhaps I'm not the only one?