A request directed to the cable itself (eg, $22 Section 3.7) could be useful just to confirm the cable is talking with Realterm. Based on earlier test results, it doesn't seem that $0D (or $0A) is required. For the cases above, these bytes are past the end of the defined message length and therefore are possibly being misinterpreted as the next (but invalid) command.
Based on my read of the manual (see Section 3.3 for an overview) the setup steps to send/receive messages with the vehicle would include verifying the OBD protocol is VPW ($31 sub $01), setting up filtering to pass back only messages directed to EDIT:$F0 ($33 sub $00), and turning on the network ($31 sub $02). Section 3.3 indicates that turning on the network with no filters set should give you essentially all the bus traffic, so that's another potential comm test.
Also, the description of the $08 network receive message seems to imply it's a wrapper of sorts around the vehicle response, vs a command that is issued to the cable. Based on Section 3.3, once the network connection is open I would expect the $08 message(s) to be generated by the cable based on received responses to requests and any filtering in effect. I'd be interested in what your testing shows.
I will try to catch up on my own testing over the next few days - hope this helps in the meantime.
Jim