1. We can keep operating the Shuttle at a cost of several hundred million dollars per mission. As the vehicle gets older, those costs would probably get higher.

2. We can try to develop some sort of replacement vehicle.

If we decide to replace the Shuttle, what course of action should we take?

2a. We could try for Shuttle II, a vehicle capable of doing everything the current Shuttle does but with newer technology.

2b. We could try for a different design altogether.

If we choose 2a, odds are we'd end up with a system that wouldn't be much cheaper to operate than the current one.

If we choose 2b, we have some flexibility. We could size the crewed vehicle for the mission of carrying a crew of perhaps 7 people to orbit, perhaps with a logistics payload of 5000 pounds or so. Large payloads like space station modules could be launched on large expendable boosters or perhaps some future unmanned reusable system.

Without some sort of crew escape vehicle, the current ISS is very marginal. Perhaps the crew return vehicle could be developed as a Block I design, designed to provide the escape capability only. As the AviationNow article I linked previously suggests, this vehicle might be launched unmanned to the ISS on a vehicle like the Delta IV. A Block II version would add the systems necessary to perform the crew launch and full orbital operations. It could share systems with the Block I CRV to some extent. The AviationNow article suggests the Block II version could be launched on a man-rated reusable booster of some sort. Perhaps that could be used to launch unmanned payloads as well.

There's another option:

3. Contract launch services to private businesses, including manned spaceflight. Let the businesses determine the best means of providing the services.

Of course, there's another option:

4. We could give up on manned spaceflight altogether.

Not that I'm advocating this option, just acknowledging it exists.