Several crystal structures of human complement factor B have been done. I don’t know if you’re looking at the human protein, or some other. If all you need to do is look at an already known structure, you can probably do that directly from the folks at PDB (http://www.rcsb.org). You’ll need the pdb file accession code which should be listed somewhere in the annotation data for the gene. I don’t know what the ncbi annotations look like, but the Swiss-Prot or ExPASY information is linked to the pdb database. Just scroll down the page until you find a section labeled 3D-Structures and there are links to the pdb for any known structures. Once you’ve pulled up the pdb or cif file, there are java applets or plug-in options for displaying the structure. Or, if you prefer, there are free pdb viewers around. Spdview (now called DeepView, I think) is available from Swiss-Prot, and cn3d (or something like that) or Chime are availble from ncbi. The latter are a little more user-friendly, while spdv is more powerful.
If you need to come up with your own homology model, that’s a bit more difficult. There is an option from the Blast output (at least there is when I run it from ExPASY) asking if you want to make a homology model. If you open that link, you will be asked to give your e-mail address, and then the server will grind through an automated alogrithm for generating what ExPASY deems is the best overall structural alignment using your sequence alignments and any known structures. The nice part is that you don’t have to do very much. The downside is that neither do you have much control over the process. But in a few hours to a day, you should get a couple of e-mails with your homology model and a summary of the aligments the server used in making the model. You get a summary file and a separate pdb coordinate file so you can display your spaghetti if you wish.