There are two things that I hear people say Apple is notably good at:
1. "Design."
2. Controlling a complicated end-to-end extremely high-quality manufacturing process.
For 1, I'm not sure I believe that there is a transferable organizational expertise between designing phones and designing cars.
For 2, I'm not sure I believe that there is a transferable organizational expertise between manufacturing phones and manufacturing cars, and I'm definitely sure that I don't believe that they can transfer organizational expertise to manufacturing cars if they purchase an extant manufacturing process/organization.
I recently drove a car with lane following and adaptive curse control and it kind of sucked. Lane following needed great road conditions so it failed miserably. But, while adaptive cruse control was nice in heavy traffic, it was also clearly tacked on. The acceleration and deceleration was felt jerky.
People can generally drive really smoothly in traffic so why suffer a machines erratic behavior when you need to focus in case it fails. Sure, do the same thing but better seems obvious but iPhone was more a response to poor completion than a truly innovative product.
Seriously, what does that look like? What are the attributes of a person who is good at integrating hardware and software on a phone, and who also is good at integrating hardware and software on a self-driving car? Your example was jerkiness in acceleration and deceleration. The people who are good at mitigating jerkiness in acceleration and deceleration are presumably mechanical engineers and roboticists/AI guys. I don't think that any of those people have been involved in Apple's flagship, reputation-making products so far.
As you allude to, it's easy to say "do the same thing but better." If all you needed was a product guy who was like, "Hey, we should make this better," then it would already be solved.
The best example I can give is stuttering when you drag stuff around. Android phones had issues for years with this because it's not really a hardware or software issue it's both. Eventually you can throw enough processing power to make this go away, but being a few years ahead of that curve is the kind of thing people will pay a premium for. Ed: It's the kind of issue you can't just point at one group and say fix this.
PS: That said, I don't think an iCar is a good idea right now. We seem to be in the Palm pilot stage of self driving cars, give it 15 years and apple could make a killing, but it's still a little early.
Now buying Ford is probably not the way to go. But, Apple can skip a lot of the 'ramp up' if they ever wanted to get into the car business.