I've never heard of Caffenol, but 18 minutes sounds really long for any developer. Check the developer data sheet again, especially noting the temperature you used. Something doesn't seem right.
Also, C-41 is more complicated than a single developing phase, and temperature needs to be well controlled.
It's not clear whether you're asking about the developing process or your exposure process. I'd separate the two and make sure one is right before trying something unusual with the other. Take some pictures with your camera, bracketing from -2 to +2 f-stops from center exposure, at maybe ½ stop increments. Get the film developed commercially by a lab you can trust.
Once you know you are exposing the film correctly, now you can try different developing techniques if that's really your purpose. It would be good to explain what you are trying to achieve that normal C-41 developing doesn't give you. Without knowing what the object is, it's hard to suggests ways to achieve it.
That all said, your negatives look under-developed, and the orange base looks too dark. However, that could be due to how the pictures of the developed film were taken, not the developed film itself. Again, getting commercially processed C-41 film will give you a reference of what it should look like.
You also say your negatives "came out as black and white". It can be deceiving to look at color negatives. Often little color is apparent just by looking at them. The orange base swamps subtle color differences. Scan the negative, correct for gray, and then see if you really have color or not.
Here is a very quick and dirty attempt to see what information is available in one of the negatives you show. It looks like a mess because the small variations in your low contrast image were magnified. But it does show some color information is present:
A proper scan with better dynamic range would do a much better job of showing you what information is in the negative.