Situation Goal testing: Multiple object IDs?

It's likely that the test is as limiting as you describe. We tend to write tuning structures that encourage the use of tags rather than definitions, simply because it's more maintainable that way (for us).