I want to lay out a schmatic and have it fabbed custom so i need to know stuff like what kind of chips and processors to use to make it work.
I do not mean this to be unkind - but you don't have a single clue as to what you are doing. Do you understand the concepts of buses, addressing, address decoding, memory mapping, peripheral interfaces, registers, and on, and on, and on...? If not, then your first trip is -not- to the parts bin, the electronics store, or the PCB fabrication facility: It is to the library (or more likely nowadays, the used/antique/rare book dealer).
If you are serious about pursuing this project, first off you need to be prepared to spend a lot of time and money. Depending on how you approach the project, the parts alone will easily outstrip the costs of a new PC, mainly because you will be dealing mostly with parts that are surplus, new-old-stock (NOS), and in some cases, collector items ($$$). Many of the parts needed haven't been manufactured in over a decade or more. You will likely need to deal with collectors who will want a lot of money for certain prized parts (if they will part with them at all). Beyond the costs, though, the amount of time needed to pursue the project will be staggeringly large as well; this isn't a weekend project. Starting from zero knowledge, it will likely take you a year (minimum) just to get up to speed on basics, and then another 1-3 years planning your architecture, and perhaps building prototypes. The "final" (in this kind of project, nothing is really "final") version will take about 6 months to a year to assemble and debug. Note that a lot of this depends on your skill with electronics and digital design; I am assumming a fairly good proficiency with my numbers. After you have something working, then you have to figure out how to get code on the thing (and an operating system? you won't get Linux or something like that on it - you might be able to get something like CP/M or Flex, or something like that running - and if you are building that from scratch - well, good luck).
You could easily dedicate a good chunk of your life to such a project - seriously; this isn't a project for the "faint of heart". If it is something you truely feel dedicated to do; a passion, if you will - then good luck, it will be a very rewarding (and trying at times) endeavor. You will gain a level of understanding of computers and the pioneers that built them in a way few do and have.
Finally - books: You are going to want to find a catalog listing, if you can, of the publisher "TAB Books" - these are long out of print; if you need, I can reccommend a few I have in my collection (I by no means have all of them or even the best of them - but I have quite a few that could be helpful - PM me). TAB was a publisher in the 1970s and 1980s of technical books; "How to Build Your Own Working Microcomputer" and titles of that nature (In fact, I believe that is one of the titles - the author's name, though, I don't remember, but I think it is in my collection). These were books that were written by a myriad of various authors, detailing all kinds of techinical topics; many of them on popular pursuits of the period: Alternative energy, lasers, hifi audio and speaker constuction, robotics, computers, software and programming, etc. They all were published with a similar style and layout - it was a very concise and detailed, yet very easy to understand and follow, for the most part. Since the style was so consistent, from book to book, if you read one TAB book, you were versed to read the others easily. Its something you don't see today (the closest I have seen has been the O'Reilly series of books).
Good luck on your pursuit.