As Thomas says, only introduce an event when needed.
Also, remember that there is no point to an event if nothing else is interested.
If you only have one list of cars, then a CarAdded event is not needed since the list can deal with the server response directly. The same is true of the other events (and maybe also StartCar, StartNewCar, CarStarted, StopCar, StopNewCar, CarStopped if I had any idea what they were for).
So in the scenario you explained, if the dialog is a popup, then since you haven't mentioned any other page, there is, as yet, no need for any events at all.
Now if you had a page for adding cars and another page listing cars (on, say, another tab) then there's a need for an event.
In an optimistic UI, when the Add page submits the 'add' you fire a CarAdded event and the list (and anything else subscribed to this event) will respond accordingly. If, subsequently, the server reports that the add failed, you'll need an undo event to. More work, but it gives a more responsive feel for the user.
In a pessimistic (traditional) UI, when the Add page, when the server reports that the add was successful, you fire a CarAdded event so that the list can respond. Easier, but the user sees a lag between pressing the button and seeing the update, and if they switch quickly to the list, their new entry might take a few moments to turn up.
If, somewhere in my app, I do something, I don't just tell the world in the desperate and pathetic hope that maybe somewhere, sometime, something or someone might take an interest in the boring, mundane minutiae of what I'm doing. An event bus is not the programming equivalent of Twitter :-) My pages are more of the busy executive, working efficiently on their own initiative as far as possible, and sending out memos on a strictly 'need to know' basis as well as responding to incoming ones. Or something like that.