Parts of your general idea are correct, others need some refinement. To skip ahead to the conclusion: I'm going to recommend that you read up about linkage disequilibrium. Everything else is background information that will hopefully help you understand why linkage disequilibrium is what I think you're looking for.
Background: Meiosis Basics
Firstly, in the bare principle, meiosis is quite simple. Let's consider meiosis in a human cell, i.e. the division of a primary spermatocyte (the only other example is division of primary oocytes). That's this one we're talking about: http://en.wikipedia.org/wiki/File:Figure_28_01_04.jpg In simple terms, the ready-to-develop primary spermatocyte contains 46 chromosomes, two (2n) each of chromosomes 1-22, an X and a Y chromosome. Two chromosomes of the same type (e.g. "chromosome 4") are called homologues. Homologues contain different versions (= alleles) of the same genes. During Meiosis I, homologues align in pairs, recombine, and then separate into two new cells (which are now 1n). The following Meiosis II is basically a mitosis, meaning that the two (identical apart from previous recombinations) chromatids of the chromosomes are separated and packaged into separate cells.
Recombination and Interstices
During the alignment of homologues in Meiosis I, the attraction between the two chromosomes depends on their homology - i.e. similarities in DNA sequences that end up being close to each other in this setting. This attraction is on the molecular level, so the tightly-condensed neighbouring DNA double-helices actually physically stick to each other because the polarities/charges across relatively large stretches of the molecules complement each other. I don't think how this exactly works is well understood yet, but what is clear is that homologues attract each other at this point because they contain mostly the same DNA sequence, apart from small differences (which make the different alleles).
How exactly recombination occurs at this point isn't entirely clear either on the molecular level (http://en.wikipedia.org/wiki/Homologous_recombination#In_eukaryotes). Just to get an idea of how it could work: while the two chromosomes are aligned, it's not difficult for them to actually tangle up and break, leaving the repair machinery with the potential to fuse the wrong chromosome ends together. The breaking and exchange might also be a programmed process with dedicated cell machinery (quite probably so I think personally).
In any case, not all locations are equally likely to recombine. It depends on levels of homology (i.e. "stickiness") between individual stretches of DNA and other unknown factors. Just in the same way, the number of break points in each chromosome isn't pre-determined.
First of all, the concept of loci is arbitrary and has no molecular basis. A locus is simply "a place on a chromosome" and could refer to an individual base(-pair) or a whole stretch of DNA. The concept is however useful when talking about inheritance, as it illustrates the principle of linkage, or rather linkage disequlibrium.
As far as I am aware, it comes from a time before we began thinking in chromosomes. One could study whether there was linkage between two traits (or more generally loci), i.e. a higher-than-random chance to inherit the two together. We now know that linkage simply means that the two loci are on the same chromosome (for example, chromosome 6). Thus, if no linkage exists, the chance to inherit both loci together is no larger than expected by random Mendelian inheritance (which makes sense as there is no molecular mechanism that makes offspring e.g. more likely than 50/50 to inherit chromosome 6 and 19 together from the same parent). If linkage exists , the loci are on the same chromosome, and any case where the two are not inherited together must be caused by recombination.
The idea of linkage disequlibrium is that loci that are close to each other on a chromosome are more likely to remain together (and be inherited together) than far-apart loci. This is obviously due to the increased chance of a recombination event occurring between them with increased distance. This actually seems to be what you are looking for as a pointer for where to go next, so I recommend using your preferred internet search engine to learn more about that :)