The old arguments evaporate in a cloud of red smokeOnly if you exclude traffic congestion as a problem. Self-driving cars can help with some optimization (i.e. closer following distance on freeways, less traffic disruption due to erratic drivers), but they aren't going solve the overriding problem of too much traffic in the city.
San Francisco is already experiencing extra congestion from the precursor to self-driving cars, car sharing [1]
And wide-spread use of self driving cars may make congestion even worse. Instead of the office worker driving to work and parking in the garage all day long, his car will drop him off,
then will go drive around and look for someone else to pick up (or will seek cheap parking outside of the city center),
so that's an extra car on the road.
And anyone that's tried to drive past a school in the morning or afternoon knows that traffic is going to be a nightmare during commute hours when every office building is surrounded by cars that are dropping off/picking up the 1000+ workers in each building.
Not to mention that your "zero-carbon" (battery-powered)" cars don't exist, even if powered by solar cells (that were created by non-carbon fuel sources), the processes that go into mining materials and manufacturing a car still emit tons of CO2 and we are a long way away from making all manufacturing carbon neutral.
Single occpancy cars are just not very efficient at moving people. A single train can carry 2000+ passengers and deliver them every few minutes. In comparison, car freeway lane carries around 1000 cars/hour.