Following the information a book titled "Road Vehicle Aerodynamics" by Scibor-Rylski - I modeled a ground effect approximation. Downside is, it doesn't take into account viscous losses and I can no longer measure lift. What you don't see in that image is the bottom half of the model, which is a mirror image. Plane 3is the virtual ground - 14 inches below the vehicle.
Why? I don't have enough memory to run even a 2D ground plane... Bugger... At low resolution, I can make it
So I've added an engine approximation thing and a radiator. The radiator is a bunch of squares (40, I think) offset from each other by about a half inch. I added my two grille openings and an outlet vent on the top (which is a bit too big methinks).
According to Scibor-Rylski - internal losses account for about 12% of total aerodrag with gains to be had up to 6%. Mind you, this book is a bit dated
-------
Is this model right? I lean towards "no." My tuft testing showed fairly steady flow the entire length of the hood. BUT, light yarn doesn't need much velocity to show a flow direction. This could very well be what's happening, but I'm not entirely convinced just yet. I'm going to measure my grille's and modify.