We agree that Civ4 definitely creates powerful incentives to have a few, highly-developed cities, then. It seems this has a lot of non-obvious implications for other aspects of play.
For example, a few highly-developed cities means that your cities will probably have large populations. Suddenly, Wonders that offer single-city effects (e.g., Copernicus' Observatory in Civ3) become much more valuable than previously, and empire-wide Wonders (e.g., Sun Tzu's Academy in Civ3) are slightly less valuable (relatively speaking), because a single large city is a much greater fraction of your nation's total output.
Another consequence of the large-city phenomenon is the change in the goals of warfare. Razing a city becomes much more costly, because of the opportunity cost of giving up a city. Pillaging improvements, however, becomes much more powerful. For one thing, large cities are more prone to starvation than small cities (more dependent on improved farmland, and less likelihood of enough food production from alternative tiles in the city radius). This will be reinforced by the fact that pillaging will now give the pillager a gold bonus.
Finally, the increased value of individual cities will probably make wars quicker. A powerful early advance that captures a city or two might grind to a halt in Civ3 (leaving aside AI incompetence). Now, the loss of a city will be a devastating blow, denying your nation a large fraction of national output and potentially causing the entire war effort to collapse.
Of course, this danger will hopefully result in more vehement defenses of individual cities, as well as encouraging the AI to sue for peace when a city's loss is imminent. Perhaps we'll see most wars take place outside of cities, and once one side reaches the enemy's walls, a peace settlement is usually reached.
