EU does have trouble with solar seasonality, but wind is seasonally anti-correlated with solar, and the geospatial correlation between different wind turbines drops off more than linearly with distance, and the EU covers a very large land mass as-is. You can also over-build solar inside Europe to have reasonable collection during winter.

I also see no reason to admit North African states into the EU before an agreement can be reached about transporting solar. The geopolitical risks have always been about other states severing the link during a conflict with you, and less about the parties to the deal reneging. So whether Morocco or Algeria is part of the EU is quite immaterial to the risk profile.

This kind of thing really does need simulation modelling to be reasoned about properly. The one thing I am confident in saying is that these single sentence just-so stories about what is and isn't a good idea are going to be wrong, because the fundamental principle is statistical diversification, which needs to be approached through simulation rather than through words.