D. Nye,
I'm no expert, but let me make the following three points:
(1) I'd be careful looking at any single picture of a race car because teams learn and adjust. For example, looking at the Mercedes side pods at the beginning of the 2022 season would send you down the wrong path. A more tangible example is the SL-C that won the national championship. The first is picture would leave you to believe that the stock location of the outlet was ideal. However, the second picture is a later version and shows a complete rethink.
Early version appears to maintain the outlet's stock location and size. Note the large Gurney flap (black piece of metal sticking up 90 degrees from the nose. Completely appropriate for a race car, but not particularly nice on a street car.
Updated version. The outlet is much larger and moved forward. There is an internal duct to guide all of the air upwards. Look at the angle of that duct and it is seems pretty clear that they were using it to generate downforce. Also note the three cutouts in left fender under the top vents to allow air to escape from the wheel well. Not shown is the massive wing with endplates that replaced the old splitter.
Note that the male buck for the SL-C was designed and constructed in clay. To my knowledge, there was no CFD performed until much later when they were chasing the national championship. To my understanding the primary outcome of the CFD was the change in the radiator duct/outlet and the Race Splitter (not to be confused with the Track Splitter which some people run on the street). The pictures below were taken from the Superlite website. The first picture shows a high pressure zone directly on top of the the stock radiator outlet which is intuitive. The second picture shows a rethink (I have no idea if it's the one that they wound up building). It looks like the vertical component of the radiator outlet flow clears the roof scoop (my CFD model indicates that the roof scoop generates a fair amount of lift).
(2) Every car is different and engineering is a game of tradeoffs. If you don't understand what their objectives and constraints were you might misinterpret things. Looking at that car, the windshield looks more raked than the SL-C's and less likely to create a high-pressure zone at the trailing edge. However, most importantly, the leading edge of the outlet is moved well forward of the stock SL-C location and the shape of the internal duct could easily be changed to move the trailing edge of the outlet forward. Would that help? I don't know and my guess is that they probably didn't either.
(3) We (as in other people) know a lot more about aerodynamics today that we did in 1967. My guess is that there wasn't enough computing power in the world to run a rigorous CFD model back then even if the software had existed. Much of the learning came from comparing CFD, wind tunnel and real-world data over the last 50 years.