This is good advice. Prior art can torpedo your application, so it makes sense to do a cursory search on your own nickel before paying a lawyer. My patent lawyers have always advised me to do my own searches.

I'm not a lawyer, but it's been my experience that if your invention combines A and B, and the examiner can find A and B in separate patents, then you're pretty much cooked. If you find both A and B in a single patent, then you'd better hope the patent is expired. If you find prior art, then your options are to give up or narrow your claims. That can be a gut wrenching, if you're already in the hole to a lawyer when you have to make this decision. At best, narrowing your claims still lets you protect the important elements of your invention.

If you find a "close" patent, then following its citations, and the patents that cite it, is a good way to chase down more possible prior art.

Now, I've found that it's touch and go whether the examiner will look outside the patent literature. As it gets easier and easier to search for anything that's been written about anything, fertile ground for finding prior art will of course expand. It could include the academic literature, so if you're a student, you might be able to get some (free) help from your friendly neighborhood reference librarian.