In a few important ways, a typical Jecklin-disk baffled omni stereo microphone setup (which is what we are talking about here, in contrast to using a baffle to modify the behavior of a single omni for other "non-stereo" reasons) is similar to near-spaced directional microphone stereo setups (ORTF, NOS, DIN, etc) in that they all represent reasonably optimized configurations designed to provide a good balance of trade-offs between contradictory aspects.
Contradictory aspects?
For example- Coincident configurations generally provide sharp directional imaging, yet often a somewhat flat and lackluster portrayal of ambience and revererance, while wide spaced configurations provide big lush ambience and an attractively open-sounding portrayal of reverberance, but with rather fuzzy, imprecise imaging. Near-spaced configurations try to find a reasonable balance between these things, while not being the best at either. Near-spaced configurations are a search for an optimal tradeoff between several important aspects which are basically in conflict with each other, trying to get each of them "right enough" simultaneously.
If you are no longer tied to a single two-channel microphone configuration designed to provide optimally "pre-baked" 2-channel stereo direct into the recorder, you can attempt to better optimize seperate microphone arrangements for one specific aspect or another, with the intent of combining them to get the best of both. Done correctly, each of the separate parts provides what it does best.. and hopefully covers the weaknesses of the others, such that their sum is greater than any of the individual parts in isolation. If not done correctly you can get a mess, and a well-optimized 2-channel microphone configuration would be better.. and simpler. Preschool analogy- We want to be careful to color between the lines rather than mush to much together into a big blob of brown fingerpaint.
Consider that we get the cleanest direct sound from a SBD patch. We get the sharpest directional imaging from a coincident pair of directional microphones. We get the most enveloping and open ambience from a pair of spaced omnis. If I have SBD or a coincident pair, it makes sense to mix in a pair of wide-spaced omnis which are most optimal for providing the ambient stuff those things lack. Any lack of close sounding clarity and fuzzy imaging from the wide-spaced omnis doesn't matter because the SBD or coincident pair provides the clear direct sound and sharp imaging. The spaced omnis are relieved of the burden of providing clear direct sound on their own or good directional imaging and even the need to provide a solid soundstage without any "hole in the middle". In fact, I can now space the omnis even farther apart than I would other wise dare, such that the portrayal of ambience and reverberance is even more open and diffuse. I wouldn't otherwise do that without having the coincident pair or SBD available because it produces a hole in the middle. But when combined with my other source(s) that hole in the middle can be an asset rather than a liability. It makes room in the middle for the clear directional imaging stuff. Likewise the coincident pair can be better optimized for clean pickup of direct sound and directional imaging (angled less wide, say PAS) without worrying about the recording sounding too monophonic and dimensionally flat. In that way such liabilities can actually be leveraged as assets, as they help keep each part from conflicting with and "stepping on" the other when they are combined.
This is the basic idea behind OMT, which seeks to optimize each part separately for a particular role while staying out of the way of the others, making combination afterward easier. And hopefully in such a way that the resulting mix works out better than attempting to find a good "pre-baked" trade-off using just two microphones that produces a mix of those aspects directly.
Lots more on this idea in the OMT threads, especially part 2.
But no need to get overly complicated. If you can record a SBD patch along with your pair of omnis, try mixing that with wide spaced omnis a few times, and with a similarly placed Jecklin-disk baffled pair a few times and decide for yourself which combination works out best.