Copy that. It certainly is valid to have tight control over the inputs (gas flow) rather that relying exclusively on feedback from the output (the temperature).If you know the exact pressure you need to be at, the easier it is to replicate a roast.
I always went to these guys for sensors and instrumentation, when I was spending OPM (other people's money). They're top notch and I always valued the online resources they offered - lots of information and advice - even if I cheaped out later and went with something else. They give you a good sense of what state-of-the-art is.
I haven't read through all of this, but maybe useful?
http://www.energy.ca.gov/2014publications/CEC-500-2014-104/CEC-500-2014-104.pdf