Hi Ken,
No need to apologize - we all had to start somewhere, and this group is for ZAP users of all levels of experience :)
I havnt used OpenVAS, so hopefully someone will correct me if I'm wrong :)
I _think_ tools like that are looking for known vulnerabilities in standard software packages.
ZAP is really targeted at custom applications where this approach falls down - there are no sets of known vulnerabilities.
ZAP will typically inject attacks into all of the input vectors it knows about and then tries to determine if they were successful.
Each type of test will work in a different ways - we try to document how they work in the help, eg
Release AscanrulesThe effectiveness of ZAP will depend on how effectively it has explored and understood the application.
Proxying via ZAP and thoroughly exploring it manually is ideal, but its manual and time-consuming. If you have a good set of regression tests (eg using Selenium) then proxying these through ZAP is also recommended (and what we do at Mozilla).
Otherwise you'll have to rely on the traditional and ajax spiders which can be very effective but also may not find everything.
ZAP also needs to understand how the application is structured. If its a 'traditional' app then you should be fine, but if its a 'single page app' and/or uses a lot of data driven content then you may well need to configure ZAP in order for it to be more effective.
Does that help?
Cheers,
Simon