Only those components that are imported in the script where the pipeline is defined would be included in the DAG plot, is that right?

Actually the way it works currently (and we might change it if there is a better way), every time you call PipelineDecorator.component a new component is stored on the Pipeline Task, which is later translated into DaG graph and Table (next version will have a very nice UI to display / edit them).
The idea is first to have a representation of the pipeline process, before it is launched, the second is to allow you to edit the DaG in the UI:
Create pipeline from code Clone in UI Edit configuration to change the pipeline parameters, for example change target execution queue for indevidual stepsMake sense ?